CRA Report On IMO 2020 June 2019
CRA Report On IMO 2020 June 2019
CRA Report On IMO 2020 June 2019
Prepared by:
Contributor:
Disclaimer
The study was commissioned by Coalition for American Energy Security (CAES). The research, analysis, results and
conclusions were all developed independently by the authors. The conclusions set forth herein are based on
independent research, use of proprietary models, and publicly available material.
The views expressed herein are the views and opinions of the authors and do not reflect or represent the views of
Charles River Associates or any of the organizations with which the authors are affiliated. Any opinion expressed
herein shall not amount to any form of guarantee that the authors or Charles River Associates has determined or
predicted future events or circumstances and no such reliance may be inferred or implied. The authors and Charles
River Associates accept no duty of care or liability of any kind whatsoever to any party, and no responsibility for
damages, if any, suffered by any party as a result of decisions made, or not made, or actions taken, or not taken,
based on this paper. Detailed information about Charles River Associates, a registered trade name of CRA
International, Inc., is available at www.crai.com.
Copyright 2019 Charles River Associates
For questions about this study, please contact Jeff Plewes at jplewes@crai.com
Page i
Economic Analysis of IMO 2020
Table of contents
1. Summary....................................................................................................................................... 1
Page ii
Economic Analysis of IMO 2020
1. Summary
IMO 2020, a regulation developed by the International Maritime Organization (IMO) to
significantly reduce sulfur emissions from the global shipping sector, will come into effect on
January 1, 2020. The regulation will require a large reduction in the sulfur content of marine
fuels consumed by most ships around the world. The resulting shift in fuel demand, from
higher sulfur fuels to IMO 2020 compliant fuels, will be supported by a well-prepared global
refining sector. The U.S. refiners are particularly well prepared for the shift as a result of
significant investments to develop complex refineries.
While marine fuels represent only about 4% of global fuel consumption, the move to
compliant marine fuels may have some impact on broader refined product prices. The U.S.
Energy Information Administration (EIA) has forecast minimal fuel price impacts of IMO 2020.
Still, consternation around IMO 2020 remains among several large fuel consuming sectors.
There has been some political pressure for the U.S. to somehow support and encourage non-
compliance with IMO 2020 (by ships) through non-enforcement (by countries).
While it is well reported that non-enforcement by the U.S. and other countries faces legal,
political and regulatory hurdles, the economic implications of a U.S.-led “defection” from IMO
2020 have not been directly studied. This study evaluates the fuel price and macroeconomic
impacts of a move to less global IMO 2020 compliance. An advanced suite of models,
including CRA’s EMR-IMO macroeconomic model and Baker & O’Brien’s PRISM refinery
model, were employed in this analysis.
The key findings related to a shift to less IMO 2020 compliance include:
Gasoline prices see no discernable change. IMO 2020 has minor impacts in both
directions on gasoline prices.
Diesel prices may change by about $0.04/gallon – less than the average monthly
change in diesel prices over the past four years. EIA estimated that diesel prices
could increase by 5% on average from 2019 to 2020, though remain below 2018
levels. With less IMO 2020 compliance, the increase could be 3% instead.
U.S. refiners, which are well positioned to see margin benefit from IMO 2020 since
they produce the fuels that will increase in demand, would be expected to see a
decrease in margin from less IMO 2020 compliance.
Of the non-energy sectors in the U.S., only the relatively small marine transportation
sector sees an economic output change greater than 0.02% in either direction.
The impact on U.S. GDP growth is not detectable at the one-hundredth of a percent
level (i.e., GDP grows at 2.16% regardless of IMO 2020 compliance levels, based on
EIA’s baseline GDP estimate and CRA’s modeling of a partial compliance scenario).
As IMO 2020 implementation approaches, there is greater clarity into how the shipping and
refining sectors will respond, and it appears the industries are driving toward a transition with
minimal price disruption or fuel availability issues. The U.S. is in a unique position, based on
its low sulfur crude production and the significant investments made by refiners, to support
the global shift in marine fuel demand.
A U.S.-led defection from IMO 2020 compliance, which is clearly a difficult, if not impossible,
proposition to execute, does not drive material benefit for the U.S. economy, with any benefits
likely accruing to other global economies instead.
Page 1
Economic Analysis of IMO 2020
2. Introduction
2.1. Background
On January 1, 2020, a regulation developed by the International Maritime Organization (IMO)
to limit sulfur in marine fuels (“IMO 2020”) will come into effect. The regulation was
announced in 2008 and has been set for implementation since 2016. It will significantly
decrease sulfur emissions from international shipping, leading to benefits for human health
and the environment.1 It will also lead to several forms of economic benefits for countries,
such as the U.S., that are best suited to support global compliance.
IMO 2020 will require a large reduction in the sulfur content of fuels consumed by most ships
around the world. This shift will be facilitated by changes in fuel production and refining,
industries that have been preparing for years to supply more low sulfur fuels to the shipping
sector while meeting all other fuel demands. Refiners in the U.S. are particularly well situated
to provide lower sulfur fuels and the crude oil produced in the U.S. is naturally lower in sulfur
and therefore will see increased demand around the world.
The pending fuel demand shifts have led to some consternation within fuel-consuming
sectors, particularly around the impacts on fuel prices in 2020. This topic has been analyzed,
both directly and indirectly, in a variety of studies, including by the U.S. Department of
Energy’s Energy Information Administration (EIA). While some price movements are
expected, the EIA does not forecast major disruptions or significant price movements. Our
review suggests the fuel production and refining sectors are expected to find economical
solutions. This conclusion is supported by the forward fuel prices at the time of this report.
Despite these assurances, there has been some political pressure for the U.S. to somehow
support and encourage non-compliance with IMO 2020 (by ships) through non-enforcement
(by countries). Non-compliance involves ships continuing to use High Sulfur Fuel Oil
(“HSFO”) while not scrubbing the exhaust gas to remove sulfur oxides. While it is well
reported that non-enforcement by the U.S. and other countries faces legal, political and
regulatory hurdles, the economic implications have not been directly studied.
Other studies to-date have focused on full implementation (with minimal non-compliance)
compared to a scenario with no implementation of IMO 2020. Such a comparison at this point
is no longer realistic since IMO 2020 will be implemented in January 2020. Therefore, a more
reasonable comparison is between full implementation and a partial implementation that
involves additional non-compliance driven by U.S.-coordinated non-enforcement.
1 “Health Impacts Associated with Delay of MARPOL Global Sulphur Standards” presented by Finland to IMO, August 2016.
2 Percentage non-compliance = (total HSFO consumed by ships not running scrubbers) / (total marine fuels consumed)
Page 2
Economic Analysis of IMO 2020
IMO 2020 studies, such as the April 2016 study by the International Energy Agency
(IEA).3
2) Partial IMO 2020 - Partial implementation assumes cumulative global non-
compliance of 30% of marine shipping fuels, allowing for greater use of HSFO. This
level of non-compliance is not a prediction of likely achievable levels, but rather an
indicative level based on one possible result of U.S.-led defections from IMO 2020
enforcement.
The following chart shows the percentages of marine shipping fuels consumed in the two
scenarios in 2019 and 2020, for both the U.S. and globally. The IMO 2020 compliant fuels
(blue) are any fuels with a sulfur content of no greater than 0.5%. The HSFO share (gray)
includes both compliant use (pre-IMO 2020 or in ships with scrubbers) and non-compliant use
(in ships without scrubbers in 2020). At the global level, in both scenarios about 15% of total
fuel use will be HSFO used in a compliant manner.
Figure 1: IMO 2020 compliant fuels as shares of marine transport fuels, 2019 to 2020
The analysis involved two advanced energy and economic models. Charles River Associates
(“CRA”) utilized a global macroeconomic model (“EMR-IMO”) to evaluate fuel price responses
to changes in compliance levels and to determine sector-level and economy-wide impacts.
Baker & O’Brien, Inc, (“Baker & O’Brien”) used its PRISM model to evaluate responses by
U.S. refiners. The PRISM model determined refinery output of various products, as well as
the refining sector’s economic outcomes in the scenarios.
Both models were calibrated to the EIA’s recently published forecasts of prices and quantities
in 2020 under full IMO 2020 implementation. To estimate the impacts of moving to partial
implementation, we modeled an increase in demand for higher sulfur fuels in the marine
shipping sector (and the associated decrease in lower sulfur fuels). The EMR-IMO model
Page 3
Economic Analysis of IMO 2020
determined the macroeconomic response, and the PRISM model determined the response by
U.S. refiners.
4 The STEO April 2019 estimates for 2020 prices were: Brent $62/barrel, WTI $58/barrel. If crude prices are higher, we would
expect both scenarios’ refined product prices to rise at about the same rate. Crude oil prices in the Partial IMO 2020
scenario are determined by the EMR IMO model and do not change materially from the Baseline IMO 2020 scenario.
Page 4
Economic Analysis of IMO 2020
Page 5
Economic Analysis of IMO 2020
U.S. refineries are well positioned to provide the types of fuels most needed in an IMO 2020
future. They currently provide over 10% of marine fuels consumed in the world, and that
share is expected to increase. Along with increased volume, refiners can expect to see their
refining margins improve with IMO 2020, though remaining within their historical range. This
is borne out in our analysis. Following the introduction of the new IMO 2020 regulations, we
see average U.S. oil refining margins in 2020 outperforming the recent past. In the Partial
IMO 2020 scenario, we see a minor reduction in product prices eroding refining margins,
compared to the Baseline IMO 2020 scenario.
Due to the option of using more of the lower-cost HSFO, the global marine transport sector
would see the greatest benefit of less compliance with IMO 2020. However, that benefit
would be mostly isolated to the trade routes for which the additional non-compliance occurs.
These benefits would be moderated by rising HSFO prices and would mostly accrue to
shipowners, as shipping rates would still largely be set by compliant vessels and the
compliant fuels that do not change significantly in price on the whole.
The overall impact on the economy is minimal. For example, under the baseline, GDP grows
2.16% from 2019 to 2020 (an EIA estimate). Under the Partial IMO 2020 scenario, GDP also
grows 2.16%. The difference is in the thousandths of a percent, which is well within the error
of the assumptions and modeling.
Page 6
Economic Analysis of IMO 2020
5 Many sources use the term LSFO for IMO 2020 compliant fuels. The distinction is important in this report due to the modeling
approach discussed later. LSFO in the traditional definition is mostly used in the power sector.
6 “IBIA offers practical advice on tank cleaning to IMO 2020 planning meeting,” IBIA, June 2018.
Page 7
Economic Analysis of IMO 2020
payback periods for investments in scrubbers. The cost of scrubbers are lower for new build
ships. Cost estimates for retrofitting existing vessels vary widely, ranging from $2 million to
$10 million. The price spread between compliant fuel and HSFO is discussed later. One
recent public analysis shows that a $200/MT price spread can lead to about $1 million in fuel
savings per year for ships with scrubbers.7
In addition to those key variables, owners evaluating the installation of scrubbers must
consider other factors. Factors supporting scrubbers include fuel diversification to hedge price
risk. Factors challenging scrubbers include concerns about the likely availability of HSFO,
which will be a secondary bunker fuel in many ports, and a rising opposition to discharge from
open-loop scrubbers in certain regions.
The amount of HSFO consumed by ships with scrubbers in 2020 is an important variable in
determining the impact of various levels of IMO 2020 compliance. To estimate the volume of
HSFO, most studies begin with estimating the number of scrubbers that will be installed
across the over 90,000 ships expected to be operational in 2020. The 2019 IEA oil report
estimated that 2,500 ships will have scrubbers by the end of 2019, 4,000 by the end of 2020,
and 5,200 by the end of 2024.8 While the longer term installation estimates seem to line up
with other industry estimates, the IEA seems to be more pessimistic than others regarding
installations prior to 2020. DNV GL reports that the total number of ships in operation and on
order with scrubbers fitted through 2019 is 2,859.9
The ships that will be fitted with scrubbers are generally much larger than average sized
vessels and have trading routes that involve more time at sea. DNB Markets estimates that
the top 5% of fuel-consuming ships (which translates to under 4,500 vessels) consume 38%
of marine shipping fuel.10 The 2019 IEA oil report estimates that the 2,500 ships with
scrubbers (plus the additional ships adding scrubbers throughout 2020) will consume 700
kbpd of HSFO, which represents about 16% of global marine fuel consumption.11 Using IEA’s
estimates, an additional 500 scrubbers could lead to the marine sector having 10% greater
HSFO demand and 7% lower MGO demand than IEA’s current estimate.
Page 8
Economic Analysis of IMO 2020
Page 9
Economic Analysis of IMO 2020
Page 10
Economic Analysis of IMO 2020
products combined.13 In the same report, the IEA estimates that the IMO 2020
regulation will lead to an additional 1.1 mbpd increase in demand for marine gasoil,
or a 3% increase in total distillate demand.
To meet such an increase in middle distillate demand, refiners can either increase
overall throughput using existing refinery setups, maximize middle distillate
production or, over a longer time period, they can increasingly employ technologies,
such as hydrocrackers, for more significant yield shift. Both approaches are likely as
IMO 2020 implementation approaches.
Producing compliant fuel oil – Refiners have already begun to test and produce
compliant fuel oils, or VLSFOs. The VLSFOs available in the market will be a range
of residual fuels that have been desulfurized through the refining process (generally
starting with lower sulfur crude) or blends of higher sulfur fuel oils with distillates.
More technically speaking, the VLSFOs will originate from “sweet crudes, de-
sulphurised fractions, cracked fractions, heavy and lighter hydrotreated fractions and
residual fuels blended down with distillates to meet the sulphur limit...”14 Therefore,
the main ways refiners will commence and then increase production of VLSFO will be
to increase the use of certain refining technologies, adjust crude inputs, and increase
distillate production (as discussed above).
Given these approaches to providing the marine shipping sector with compliant fuels, the
options for refiners fall into five main categories, as listed below.
1) Increasing overall throughput – This option could have the most impact on other
refined product markets beyond compliant fuels, with price impacts in both directions.
It can increase supply of other refined products, which would tend to decrease their
prices. However, there is an upward pressure on crude prices given the increased
overall demand.
The extent to which increased throughput is an available option for each refinery
depends on the pre-2020 utilization and the total capacity. Total capacity by refinery
has proven to be an imprecise estimate of potential. Many refineries can run above
their rated capacities for limited periods of time. In fact, in the U.S., entire PADDs
have had multiple weeks of utilization over 100% since 2014 (7 weeks in PADD 2
and 2 weeks in PADD 4).15 Over the longer term, even in the absence of substantial
investment in new capacity, the U.S. refining sector continues to debottleneck
facilities resulting in small increases in capacity (known as capacity creep).
Another possible constraint is the potential over-production of high sulfur fuel oils
without adequate disposal options. A significant portion of HSFO that will no longer
be used in the shipping sector is expected to be absorbed in the global power sector,
or used in carbon black or asphalt production. Additional HSFO production can also
be used as an input in refineries, particularly by refineries that may use these fuels as
an alternative to heavy sour crudes, like Venezuelan crude, that has been in short
supply.
14 “Detailed information on preparation and operation on fuels with maximum 0.50% sulphur", MAN Energy Solutions, March
2019.
15 “Refinery Operable Capacity Percent Utilization,” U.S. EIA, May 8, 2019.
Page 11
Economic Analysis of IMO 2020
2) Changing operations while using existing infrastructure – The refining sector has
repeatedly proven creative and resourceful in adapting to demand and fuel quality
changes in the industry. There is significant market uncertainty as to the extent
operational and other small changes can alter product yields. As discussed, there are
multiple degrees of freedom for refiners to adjust operations and change yields. With
the industry anticipating demand for compliant fuels, operational changes are already
underway. Many changes associated with improving their operating flexibilities (such
as additional tankage to segregate crude oil and fuel oil products) will be made during
maintenance activities throughout 2019.
3) Utilizing certain refineries more – It is commonly understood that the refinery-level
impacts of the IMO 2020 regulation will vary depending on refinery type. Generally,
more complex refineries with more heavy fuel oil upgrading technology (such as
cokers) are likely to benefit and target increasing production. These refiners are able
to process and upgrade heavy sour crudes into refined products (gasoline and
distillates), while producing minimal fuel oil products. They are also able to upgrade
high sulfur oil produced by other refineries. Therefore, they are likely to benefit from
improved distillate margins/prices and heavy sour crude oil price discounts, and lower
prices for high sulfur fuel oil. The U.S. has a higher concentration of these refineries,
often referred to as “deep conversion” refineries, than the rest of the world.16
Conversely, “simple” refineries that have limited heavy fuel oil upgrading capacity
generally produce more fuel oil for sale, and as a result are more exposed to high
sulfur fuel oil price changes, and potentially may be forced to switch to light sweet
crudes. These refineries are more likely to cut rates either to reduce high sulfur fuel
oil production or accommodate an alternative light-sweet crude oil. The refineries
most likely to see negative impacts are these “simple” refineries, particularly those
processing higher sulfur crudes and unable to adjust crude inputs or to make the
more in-demand fuels. These types of refineries are more common in Europe and
Asia.
4) Adding conversion and treating technologies to change yields – Refiners have known
about IMO 2020 for many years. They have had a decent amount of certainty of its
implementation for at least four years. In that time period, many refineries have
added desulfurization, coking or hydrocracking capacity that would reduce the
amount of HSFO they produce and/or increase volumes of the fuels needed for IMO
2020 compliance.
5) Altering crude inputs – Many refineries can alter the type of crudes they process
based on price spreads and availability of the various crude types. There will likely be
economic incentive for many simpler refineries to shift to sweeter crude oil (such as
U.S.-produced crude oil, see Section 3.3.2), thus improving the production of
compliant fuels and/or distillates that will see higher demand. With the curtailing of
the marine bunker fuels market as an outlet for heavy, high sulfur residues, intra-
refinery trade of these intermediates – from simpler to complex refineries – is likely to
increase. For U.S. refineries this provides potentially an alternative feedstock to
sanctioned Venezuelan heavy oil for maximizing the utilization of their heavy oil
upgrading units, such as cokers.
Each of the refinery options are interrelated, and all are likely responses to IMO 2020 in
varying degrees. There are many indications that refinery responses are already underway.
This is indicated by many announcements by refiners that they are already testing and will be
16 IEA, “Oil Refineries,” Energy Technology Systems Analysis Programme, April 2014.
Page 12
Economic Analysis of IMO 2020
prepared to offer compliant fuels well before IMO 2020 implementation on January 1, 2020.
=For example, the following refiners, among others, have made such public announcements:
Shell,17 Exxon Mobil,18 Chevron,19 Marathon,20 and BP.21
In sum, oil refiners in the U.S. and elsewhere have successfully met previous clean fuel
regulatory tightening with a smooth and coordinated transition. They have seen IMO 2020 on
the horizon and have a multitude of operation and investment “levers” at their disposal to
respond to its implementation before the end of this year. As a result of significant
investments to develop complex refineries, the U.S. refiners are particularly well prepared
and set to benefit.
17 Shell: https://www.shell.com/business-customers/marine/imo-2020.html
18 ExxonMobil: https://www.exxonmobil.com/en/marine/technicalresource/press-releases/2020-compliant-fuels
19 Chevron: https://www.reuters.com/article/us-chevron-shipping-idUSKCN1T61G7
21 BP: https://www.bp.com/content/dam/bp/business-sites/en/global/bp-global-energy-trading/documents/what-we-
do/marine/marpol-technical-guide-oct-2018.pdf
22 Cheong and Cang, “A Fleet of Tankers Is Hoarding Oil for a Gathering Storm,” Bloomberg, April 29, 2019.
Page 13
Economic Analysis of IMO 2020
In addition to the EIA, and with the exception of a few studies, the market seems to agree that
significant refined product price increases before or during January 2020 are unlikely, as well
as in the months following January 2020 (though trade volume is extremely low that far in the
future). This can be inferred by reviewing commodity futures prices for refined petroleum
products. Diesel and gasoline are the refined products with the highest product consumption
and futures trade volumes. Most other refined products, such as heating oil, jet fuel and
bunker fuels, are not traded forward at a high volume, and instead they can be hedged
through diesel and gasoline futures, or through crude futures or forward contracts. While
there are also compliant bunker fuel futures contracts available, they are not yet traded at a
volume to be informative.
The following chart shows futures prices for diesel (New York Harbor Ultra-Low Sulfur Diesel)
and gasoline from July 2020 through the end of 2020.
Figure 4: Diesel and gasoline futures prices (cents/gallon), July 2019 to Dec 2020
Source: Bloomberg
Diesel prices see a gradual price increase prior to IMO 2020 implementation, but the price
increase is only $0.04/gallon (<2%) over six months. For reference, the average absolute
change in diesel prices per month from 2015 to June 2019 was nearly $0.10/gallon.23 Also,
the increase in futures prices follows a general expectation that New York Harbor diesel
prices increase in the winter due to Northeast U.S. heating demand.
Gasoline futures prices show even less of an expected impact. As discussed later, there are
aspects of IMO 2020 response that could push gasoline prices in either direction, and the
futures show no significant upward or downward price movement near January 2020. Rather,
there is a consistent downward trend that reflects crude price expectations.
There have been a series of articles suggesting that these futures prices are not yet capturing
expectations for price impacts of IMO 2020. Such an assertion runs counter to the
fundamental theories of efficient markets and the economics of futures prices. Simple logic
Page 14
Economic Analysis of IMO 2020
can explain why – if fuel consumers (or speculators) believe there is a likelihood of a spike, it
would make sense to secure future pricing at the low current futures prices, at the very least
as a hedge. As such purchases are made, the price would increase in response to the
demand. Eventually, an equilibrium price would be found at the market’s collective view of the
most likely price at the future delivery date (with minor adjustments for risk and asymmetric
information). Any significant deviation would be an arbitrage opportunity in a liquid market. If
the market believed the studies that estimate price impacts of 20% or more, the futures prices
would certainly reflect that by now.
Source: eni
Page 15
Economic Analysis of IMO 2020
The IMO 2020 regulation will benefit countries that produce light sweet crudes, such as the
U.S., and will negatively impact producers of heavy sour crudes (such as Saudi Arabia,
Russia, Iraq, Iran, Venezuela, and Canada). Therefore, the opposite can be said of a move to
Partial IMO, as the higher demand for HSFO and lower value of low sulfur fuels lead to a
decrease in the value of U.S. crude oil.
The global fuels markets are already experiencing a “lightening” and “sweetening” of crude
oil. This is mostly driven by the expansion of U.S. production. More recently, there have been
significant production cuts by several countries with more sour crudes, such as Iran and
Venezuela.25 Interestingly, this has the impact of making HSFO scarcer since there is less
residuum from processing heavy sour crudes. This will likely decrease impacts of IMO 2020
since it represents a crude-driven reduction in the spreads between HSFO and other marine
fuels.26
25 Xu. Conglin. “Crude quality imbalances,” Oil and Gas Journal, March 26, 2019.
26 “How the bottom of the barrel became the top of refining margins,” International Shipping News, May 6, 2019.
27 “The Effects of Changes to Marine Fuel Sulfur Limits in 2020 on Energy Markets,” EIA, March 2019.
Page 16
Economic Analysis of IMO 2020
In the Partial IMO 2020 scenario, U.S. consumption of HSFO is allowed to significantly
increase. It is only constrained by the mostly fixed demand for lower sulfur fuels consumed by
ships requiring compliant fuels (for reasons such as complying with Flag State requirements
or heading to destinations in Port States that enforce IMO 2020 compliance). In the Partial
IMO 2020 scenario, HSFO consumption in the U.S. reaches 40% of marine fuel sales,
compared to only 5% of bunker sales in the Baseline IMO 2020 scenario. This includes
HSFO sales to both scrubber-less ships (non-compliant) and ships with scrubbers (possibly
compliant). The global and U.S. shares of marine fuel volumes are shown in the figure below.
Page 17
Economic Analysis of IMO 2020
The following sub-section describes the analysis supporting the definition of the Partial IMO
2020 scenario.
29 UNCTADSTAT, “Merchant fleet by flag of registration and by type of ship, annual, 1980-2018.” United Nations Conference
on Trade and Development.
Page 18
Economic Analysis of IMO 2020
Flag States could encourage non-compliance either by refusing to fine vessels or, more likely,
by staying somewhat compliant with IMO 2020 by levying extremely low fines. However, Flag
States’ control over compliance is limited since Port States can enforce compliance on any
ship entering their ports and waterways. It is therefore highly unlikely that significant levels of
non-compliance would be driven by Flag State non-enforcement.
Port States – Driving non-compliance through Port States would require a more blatant dis-
regard for the IMO 2020 rules, particularly after March 2020. As of March 1, 2020, ships with-
out scrubbers are not permitted to carry non-compliant fuels, meaning that Port States allow-
ing fuel suppliers to sell HSFO to ships without scrubbers would be directly disregarding the
MARPOL regulation.30 Assuming the Port States were to comply with that regulation, their
only option for encouraging non-compliance would be offering lax or non-existent monitoring
of ships entering their ports and waterways. This could allow both scrubber-less ships to ar-
rive with non-compliant fuels and ships with scrubbers to arrive without proof of having used
their scrubbers.
Even if the U.S. was willing to violate its IMO 2020 obligations as a Port State and allow ships
without scrubbers to bunker with HSFO, it would still need to coordinate with other Port
States that would commit to not enforcing IMO 2020 regulations on incoming vessels. Even
then, such an arrangement alone would only have a minor impact on global non-compliance
given the U.S. position in global bunker fuel sales and the high percentage of ships leaving
the U.S. for countries that would not be willing partners in a non-compliance scheme.
According to the IEA, the U.S. is responsible for about 7% of global bunker fuel sales. This is
shown in the chart below, which was provided in a report by asset manager Shroders.31 The
30 “Implementation of sulphur 2020 limit - carriage ban adopted,” IMO, October 26, 2018.
31 Odey and Lacey, “IMO 2020 – Short-term implications for the oil market,” Schroders, August 2018.
Page 19
Economic Analysis of IMO 2020
OECD estimates that the “OECD Americas” region, which includes the U.S., will represent
just 12% of global bunker fuel demand in 2020.
Figure 8: Country shares of global bunker fuels (IEA, Shroder)
Of the Port States with significant shares of global bunker sales, most have already commit-
ted to enforcement of IMO 2020. For example, Singapore (28%) has threatened prison sen-
tences of up to 2 years for captains of ships that violate the regulation. The Port of Rotterdam
(10%), the largest port in Europe, directly lobbied for IMO 2020.32
Even countries that may not have been expected to enforce IMO 2020 have signaled intent to
enforce compliance. For example, the UAE (14%) is not a member of MARPOL, yet its glob-
ally significant port, the Port of Fujairah, stated publicly “Ships will have to use compliant fuel
once the IMO 2020 sulphur cap comes into force.”33
There are a few port countries in the above chart that would not be expected to enforce IMO
2020 rigorously. For example, Russia (6%) directly challenged the IMO decision on not allow-
ing vessels without scrubbers to bunker with non-compliant fuels. However, even in these
port countries with potentially lax enforcement, a significant share of bunker sales will likely
be compliant fuels, since ships heading to ports with enforcement will demand such fuels.
Regardless of any decisions by individual or multiple port countries to reduce enforcement,
there are other forces that could drive compliance:
Carriers and shippers - A large number of carriers have announced plans to comply
and would likely still choose compliance even when given the opportunity to not com-
ply. Additionally, there are many large shippers that may require their contracted car-
riers to comply to meet sustainability goals and avoid negative customer perceptions.
Page 20
Economic Analysis of IMO 2020
34 https://www.marsh.com/content/dam/marsh/Documents/PDF/UK-en/the-sulphur-conumdrum.pdf
Page 21
Economic Analysis of IMO 2020
It is clear that the U.S. would need to work with a large number of countries to form a coalition
of IMO 2020 non-compliance trade partners in order to greatly increase HSFO sales from the
Baseline IMO 2020 scenario. The EIA estimates the HSFO share of U.S. bunker fuel sales
will drop from 57% in 2019 to 2% in 2020. Given updates in scrubber uptake, we assume
HSFO only drops to 5% in 2020 in the Baseline IMO 2020 scenario. Under the Partial IMO
2020 scenario, we assume it rises again to 40% of bunker sales, representing a fairly
aggressive view on the number of countries with which the U.S. would coordinate non-
compliance. For context, 40% of U.S. bunker fuel sales represents only about 4% of global
bunker fuel sales.
5. Modeling Methodology
The modeling framework for this study was designed to capture a variety of impacts of
changing global IMO 2020 compliance levels. The impacts of a significant change in sulfur
content in marine fuels will be mostly driven by the complex responses in the marine shipping
and fuel refining sectors. If those sectoral responses cause shifts in fuel and shipping prices,
the overall impacts can flow throughout the global economy. Therefore, to capture IMO 2020
compliance impacts, a modeling framework must include global and U.S. macroeconomic
impacts by sector, with adequate representation of the most directly impacted sectors.
To capture the range of impacts, this analysis involved two advanced energy and economic
models. CRA utilized a global macroeconomic model (“EMR-IMO”) to evaluate fuel price
responses to changes in compliance levels and to determine sector-level and economy-wide
impacts. Of all sectors of the economy, the most important to understand in evaluating IMO
2020 impacts is the refining sector. In coordination with CRA, Baker & O’Brien used its
PRISM model to evaluate responses by U.S. refiners, based on fuel price changes
determined in EMR-IMO. The PRISM model determined refinery margins and output of
various products, which greatly informed the assessment of refinery sector impacts of the
scenario.
Page 22
Economic Analysis of IMO 2020
Both models were calibrated to the EIA’s recently published forecasts of prices and quantities
in 2020 under full IMO 2020 implementation.36 To estimate the impacts of moving to partial
implementation, we modeled an increase in demand for higher sulfur fuels in the marine
shipping sector (and the associated decrease in lower sulfur fuels). The EMR-IMO model
determined the macroeconomic response, and the PRISM model determined the response by
U.S. refiners. The following chart shows the modeling framework:
Figure 10: Modeling Framework Diagram
36 As mentioned previously in this report, “full” implementation includes an assumed baseline level of non-compliance.
37 https://www.eia.gov/outlooks/aeo/
38 https://www.eia.gov/analysis/pdfpages/coal_powersectorfuelcostsindex.php
Page 23
Economic Analysis of IMO 2020
2019 (“STEO January 2019”) was the first STEO that included months within the year
2020, and thus explicitly included IMO 2020 implementation. The most recent version
of the STEO at the time of the modeling for this study was April 2019 (“STEO April
2019”).
IEO – The IEO “focuses on how different drivers of macroeconomic growth may
affect international energy markets in three heavily populated and high economic
growth regions of the world: China, India, and Africa.” The latest version, IEO 2018,
was published in July 2018 and did not explicitly model IMO 2020. It was only used
for the global model calibration.
IMO 2020 special report – In March 2019, EIA released a special report “The Effects
of Changes to Marine Fuel Sulfur Limits in 2020 on Energy Markets.” This report was
mostly based on the AEO 2019 modeling, but it included some additional information
relevant to understanding EIA’s view of domestic marine fuel consumption.
It is important to note that these studies informed the Baseline IMO 2020 scenario, but the
models were used to determine the results in the Partial IMO 2020 scenario. The results can
be reasonably applied in an environment with small changes in EIA’s 2020 forecasts since
most of the model results are initially calculated as percentage changes from the baseline.
5.2. EMR-IMO
Numerous studies aim to quantify the first-order impacts of the IMO 2020 regulation on the
global refining and marine transport sectors. These studies generally include isolated
modeling of either or both sectors, often basing future changes on past price and demand
statistical relationships without multi-sector equilibrium. Such constructs can lead to extreme
results, such as balancing the diesel fuel market by forcing expected diesel prices to levels
considered necessary to “destroy” demand in other sectors in response to additional demand
in the marine transport sector. In addition, these studies do not incorporate the many different
sectors of the economy that form integrated and complex responses to changes, such as
IMO 2020 implementation.
It is therefore preferable to review the potential change in compliance with IMO 2020 using a
computable general equilibrium (CGE) model of the global economy. CRA’s Energy and
Macroeconomic Response (EMR) is a static CGE model generally used for evaluating policy
and other changes in energy markets. For the purpose of this study, CRA has developed
EMR-IMO, a version of EMR with detailed representations of the petroleum products and
marine transport sectors. EMR-IMO uses the Global Trade Analysis Project (GTAP) database
as its core data and calibrates the base year’s (2019) benchmark data to EIA’s STEO April
2019 and AEO 2019. The state of the economy in the year 2020, including GDP growth and
prices (and demand) of petroleum goods under the IMO regulation, is also calibrated to EIA’s
STEO April 2019 and AEO 2019 reports.
5.3. PRISM
Baker & O’Brien applied their PRISM™ (“PRISM”) refining and marketing supply chain
modelling system for a more in-depth analysis of the impact of IMO 2020 compliance
changes on U.S. refineries. PRISM is widely used in the refining industry and by government
agencies. The database includes models of all U.S. refineries. Each calendar quarter, Baker
& O’Brien collates industry data, and then performs a detailed analysis of each refinery,
estimating its crude slate, product yields, and fixed and variable operating costs. The end
result is that each quarter Baker & O’Brien produces an estimate of the financial performance
of each U.S. oil refinery. PRISM is an established and proven refinery modelling platform that
is grounded in industry data. Therefore, it provides a reliable and strong platform for studies
of this nature.
Page 24
Economic Analysis of IMO 2020
Baker & O’Brien has been completing its U.S. quarterly refinery assessment since the late
1990s. Consequently, this legacy, coupled with the direct refinery industry experience of the
Baker & O’Brien consulting team, provides Baker & O’Brien with a detailed understanding of
the U.S. refinery system.
The standard PRISM refining and marketing supply chain database contains quarterly models
for over 120 refineries. Analyzing each individual U.S. refinery is beyond the scope of this
study. Therefore, Baker & O’Brien selected nine representative refineries that are
representative of the U.S. PADDs and the U.S. as a whole. This allowed analysis of
responses by different types of refineries in different regions, and the ability to aggregate
results to determine industry-level outcomes.
The Baker & O’Brien methodology focused on selecting a representative pair of refineries for
each PADD. The weighting between the refineries was adjusted to best match key
parameters – such as variable margins or product yields – for the PADD and the U.S. overall.
The following diagram illustrates the overall methodology for using PRISM in this study:
Figure 11: Overall PRISM-related Methodology
As mentioned previously, the model was calibrated to STEO April 2019. U.S. refined
petroleum product production data is not explicitly available in the STEO and was therefore
calculated by Baker & O’Brien from the STEO, Table 4a (U.S. Petroleum and Other Liquids
Supply, Consumption, and Inventories). The target yield profile for the Baseline IMO 2020
scenario was obtained by applying the STEO U.S. refined product yield shifts (from 2018 to
2020) to the yield profile for PRISM’s baseline 2018 scenario for the nine representative
refineries.
To assess each of the scenarios with PRISM, a detailed crude oil and petroleum product
price set is required. For the Baseline IMO 2020 scenario, the underlying spine of the price-
set, in terms of marker crude oil prices (such as Brent) and U.S. refined product prices,
originates from STEO April 2019 and includes a variety of assumptions made by Baker &
O’Brien. For the Partial IMO 2020 scenario, changes in price sets were determined by EMR-
IMO modeling
Page 25
Economic Analysis of IMO 2020
6. Results
This section highlights key results of the EMR-IMO and PRISM modeling.
Diesel prices – Diesel includes marine fuels, but the majority is used in other sectors (over
94% in other sectors globally, over 96% in the U.S.). In the Baseline IMO 2020 scenario,
which is calibrated to EIA forecasts, average diesel prices increase 5% from 2019
($2.03/gallon) to 2020 ($2.13/gallon). In the Partial IMO 2020 scenario, average diesel prices
increase 3% from 2019 ($2.03/gallon) to 2020 ($2.09/gallon). Neither of these increases is
39 The STEO April 2019 estimates for 2020 prices were: Brent $62/barrel, WTI $58/barrel. If crude prices are higher, we would
expect both scenarios’ refined product prices to rise at about the same rate. Crude oil prices in the Partial IMO 2020
scenario are determined by the EMR IMO model and do not change materially from the Baseline IMO 2020 scenario.
40 Refiner prices, as defined by EIA = Total revenue derived from the sale of product during the time period divided by the total
volume sold; also known as the weighted average price. Total revenue should exclude all taxes but include
transportation costs that were paid as part of the purchase price.
Page 26
Economic Analysis of IMO 2020
significant relative to historical changes in annual diesel prices. For example, between 2017
and 2018 the average price increased over 25%. In both scenarios, average diesel prices in
2020 remain under 2018 levels. This story is mirrored in heating oil prices, which are almost
perfectly correlated with diesel prices.
The 5% diesel price increase in the Baseline IMO 2020 scenario is also not entirely caused
by IMO 2020 implementation. There are two primary drivers of diesel price increases
unrelated to IMO 2020. The first, as mentioned, is crude oil prices. For example, the previous
annual EIA forecast, AEO 2018, did not model IMO 2020 implementation, yet it estimated a
14.2% increase in diesel prices from 2019 to 2020. This larger increase was mostly driven by
AEO 2018’s estimate of a 26% increase in crude price.41 The second, related driver of higher
diesel prices is increased demand due to economic growth.
Gasoline prices - Motor gasoline prices would be expected to see almost no benefit to
reduced IMO 2020 compliance. This appears to be the result of: 1) a balance between
competing impacts of IMO 2020, and 2) a gasoline market large enough to absorb small
changes without material price disruptions. The competing price impacts in response to less
IMO 2020 compliance include:
Upward price pressure – As discussed in Section 3.2.2, an expected refining sector
response to IMO 2020 is increased overall throughput to produce more middle
distillates. This additional crude processing will lead to increased gasoline production.
Therefore, one impact of a decrease in IMO 2020 compliance would be slightly lower
gasoline supply, and therefore an upward pressure on gasoline price.
Downward price pressures – With less demand for diesel products, refinery
economics would slightly shift back to supporting more production of gasoline. In
addition, some intermediates, such as vacuum gasoils (VGOs) which are generally
often converted into gasoline blendstocks, would see less demand for contributing to
IMO 2020 compliant fuel production.
These gasoline price drivers were not explicitly modeled and are not expected to have a
material impact on overall prices.
41 AEO 2018
Page 27
Economic Analysis of IMO 2020
The following chart shows the 2018, 2019 and 2020 refiner prices for the modeled fuel oils.
Figure 13: Fuel oil prices, refiner prices (cents/gallon), 2018 to 2020
The Partial IMO 2020 scenario leads to a shift in marine shipping demand from lower sulfur
fuels (including VLSFO) to HSFO. This leads to lower VLSFO prices and higher HSFO prices
than in the Baseline IMO 2020 scenario. The main beneficiaries of lower VLSFO prices will
be shippers and consumers in regions of the world that remain in full compliance with IMO
2020. In the Partial IMO 2020 scenario, nearly 40% of marine fuels sold in the U.S. would be
HSFO, thus the HSFO price increase may dampen some of the economic benefit of non-
compliance.
Petroleum product prices are far more sensitive to changes in crude market developments
than to changes in IMO 2020 implementation levels. For example, without IMO 2020 in effect,
diesel prices dropped nearly 19.8% from January 2018 ($1.85/gal) to January 2019
($1.48/gal), mostly driven by a nearly identical (19.3%) drop in WTI crude price.
Page 28
Economic Analysis of IMO 2020
GDP growth rate. More specifically, the U.S. GDP grows at 2.16% from 2019 to 2020 in both
scenarios.42
Increased refining margins are not only beneficial to refining companies, they are also
indicative of a healthy industry that contributes to the economy in many ways. In particular,
the EMR-IMO modeling sees labor employment in the fuel manufacturing sectors experience
substantial growth in the Baseline IMO 2020 scenario. This is important since the industry
provides significant employment at relatively high wages. According to the U.S. Bureau of
Labor Statistics (BLS), in 2018 employment in petroleum refineries and other lubricating oil
manufacturing sectors amounted to over 80,700 jobs. These are high wage jobs, with an
42 This GDP growth level is calibrated in the Baseline IMO 2020 scenario to the STEO April 2019 GDP estimates.
Page 29
Economic Analysis of IMO 2020
average annual wage of nearly $134,200. As a point of reference, the BLS reported a 2018
national annual mean wage of about $57,200.43
According to extractions from BLS figures, employment in the refining sector grows by 1.33%
from 2018 to 2019.44 EMR-IMO results indicate that, under the Baseline IMO 2020 scenario,
growth in employment increases significantly to 3.46% between 2019 and 2020. Some of this
growth is attributed to GDP growth. However, considering the STEO April 2019 projected
higher GDP growth between 2018 and 2019 (2.44%) than between 2019 and 2020 (2.16%), it
is reasonable to consider the IMO 2020 regulation as a reason the refining sector
employment growth outpaces overall economic growth. The figure below shows historical and
forecasted employment in the refining sector.45
Figure 15: Employment in Petroleum Manufacturing Sectors in the Baseline IMO 2020 Scenario
43 Bureau of Labor Statistics data in this paragraph sourced from the May 2018 Occupational Employment Statistics database.
Data for NAICS industry codes 324110 (petroleum refineries) and 324191 (petroleum lubricating oil and grease
manufacturing).
44 Bureau of Labor Statistics, Current Employment Statistics survey.
45 The figures in this chart also include all jobs associated with NAICS 324, “petroleum and coal products manufacturing.”
Historical data was not available for petroleum refineries only, and that definition is also too narrow to capture refining
employment impacts associated with IMO 2020.
Page 30
Economic Analysis of IMO 2020
46 NAICS industry codes 211 (oil and gas extraction), 213111 (drilling oil and gas wells) and 213112 (support activities for oil
and gas operations) are considered in the oil and gas extraction sectors.
47 Bureau of Labor Statistics data in this paragraph sourced from the May 2018 Occupational Employment Statistics database.
Page 31
Economic Analysis of IMO 2020
7. Conclusion
This report showed that a U.S.-led defection from IMO 2020 compliance would be a difficult, if
not impossible, proposition. The U.S. is not a significant flag state for ships and non-
enforcement at U.S. ports, which sell 10% or less of global bunker fuels, is limited in
effectiveness since most ships would demand IMO 2020 compliant fuels regardless of U.S.
enforcement prospects. A significant decrease in IMO 2020 compliance would therefore
require significant coordination with other countries, though most countries and major ports
have committed to enforcement.
This study determined that such an effort, even if successful in reducing global IMO 2020
compliance, would bring the U.S. no discernable fuel price or macroeconomic benefit. The
most significant impact would be on the cost of marine fuels for ships that would no longer
comply with IMO 2020, but that benefit is not concentrated in the U.S. The impact on diesel
prices is less than the average historical monthly fluctuation in diesel prices.
The U.S. is well positioned to support the global shift to lower sulfur marine fuels, both at the
refinery and crude production levels. Global refiners and shippers have had many years to
prepare, and it appears the industries are driving toward a transition with minimal price
disruption or fuel availability issues. The results of this economic analysis, which do not even
include the expected environmental and health benefits, support U.S. compliance with IMO
2020.
Page 32