THIS DISCUSSION PAPER CONTRIBUTES TOWARDS THE FOLLOWING SUSTAINABLE DEVELOPMENT GOALS:

important note: you can click on the graphs to expand them. This will also be mentioned for each graph within the main text

EXECUTIVE SUMMARY:

In this discussion paper, we look at the potential effects of adopting electric cars and their outcomes on the Kingdom of Bahrains Energy consumption and CO2 emissions. We construct a simplified model that looks at different potential adoption rates of electric vehicles and how this affects transportation energy and emissions outcomes based on assumptions about fuel consumption, internal combustion engine (ICE) vehicle fuel efficiencies, and other essential variables. Under the rapid adoption case with a ban on new sales of ICE vehicles in 2040, the car stock would end with 37% of all cars being electric (608,000 vehicles) in 2050.

With 37% of the car stock being electric vehicles, we see that the Kingdom would see 17% lower energy usage for the transportation sector, 0.84% lower overall energy usage, 8% reduction in per year emissions for the transportation sector and a 0.51% reduction in overall emissions per year. However, it is important to note that the model is likely to significantly underestimate the savings potential due to the limitation of data and the observed trends, which in turn significantly shrinks the percentage of energy and emissions from the transportation sector. We discuss the issues and potential remedies for our current model in detail.

With our results, we recommend that the Kingdom of Bahrain looks at reducing the total cost of ownership for BEVs, introducing appropriate infrastructure for an electrified transportation system, introducing benefits for BEVs gradually, and also introducing progressive policies that increases the total cost of ownership for current ICE vehicles, with a subsequent ban of all new ICE vehicle sales by 2040.

INTRODUCTION

Since the discovery of oil in 1932, the transformation of the economy and living standards of the average Bahraini has been expeditious. As a result, the Kingdom had transformed from what was once a small and localized economy of agriculture and pearl diving into that of finance & banking, manufacturing, transportation & communications, and various other non-oil-based sectors (Ministry of Information Affairs, n.d). However, the rise in living standards and wealth also lead to increased emissions due to the high usage of hydrocarbon fuel sources. Currently, Bahrain is the 7th highest emitter of emissions per capita in the world, with CO2 emissions of 20.55 metric tonnes per capita (The Global Carbon Project, 2020).

The use of hydrocarbons as the primary energy source presents two issues in the environmental-energy economics space. First, the use of hydrocarbon-based energy sources entails large emissions, meaning that there are environmental damages in the form of not only carbon emissions (a driver of climate change) but also other emissions and environmental damages that have high externality costs towards the Kingdom. Second, the use of hydrocarbons is constrained by the amount available and entails an opportunity cost in terms of internally consumed hydrocarbons that could potentially be re-exported or reduce the total amount of investments required for hydrocarbon extraction due to domestic demand for such resources.

One promising avenue to reduce overall energy use is potentially electrifying the Kingdoms transportation. Electric vehicles are more energy-efficient, where over 77% of the electrical energy is converted towards power at the wheels (EPA, n.d). This figure sits between 12 to 30% of conventional gasoline cars. Therefore, what are the potential effects of electrifying Bahrains transportation sector? Can electric mobility reduce energy usage? Furthermore, can electric mobility also reduce carbon emissions and thus reduce the externality cost they impose on the economy? We attempt to answer these questions in this discussion paper.

We look at building a simple model for measuring the potential energy usage and emissions changes with the introduction of electric vehicles in Bahrain. The discussion paper uses trends and assumptions around key data points to answer changes in energy consumption (represented in Terra-joules) and emissions (specifically CO2 consumption). We will go through the following sections of the discussion paper. The first is defining our key variables and assumptions, the second is the main results, the third is a literature review & potential policy recommendations, the fourth is the limitations & potential remedies for our model, and the last is our conclusion.

Defining Key Variables and Assumptions

Here we are going to define the key variables and assumptions made in the calculations of the results of this discussion paper; the sections will be broken down into the following; car stock, fuel consumption & estimated total km driven, transportation emissions, and electricity basted statistics. Alongside each defined variable is a reference to the appendix, which has key equations for its calculations.

Car Stock (Appendix A)

To begin defining our key variables and assumptions, we will start with our car stock. Our Car stock will take the form of either internal combustion vehicles (ICE) or Battery electric vehicles (BEV). Our data is from the General Directorate of Traffic/Ministry of Interior, which spans from 2004 to 2020 (GDT & MOI, 2020). We assume that all vehicles in Bahrain are ICE and that there are currently no BEVs. The reason why we make this assumption is because of the lack of data available on the breakdown between ICE, Hybrids, or BEV vehicles that are present in the Kingdom. Furthermore, given that the very first charging station had only been inaugurated in 2021, this indicates that the adoption of BEVs had only been recent and thus is unlikely to be a significant number of vehicles in the Kingdom (bna, 2021).

Page 1

From 2021 onwards, we assume that the total car stock will take the form of a linear trend, where in 2050, the final total registered vehicles (stock) will be 1.6 million vehicles (starting from an initial value in 2020 of 757 thousand vehicles under our linear trend). This means that each year, the Kingdom will see a net increase of registered cars of around 29 thousand. From here, we will assume six different adoption rate assumptions; Low, low-medium, medium-high, high, and Rapid+Ban of all new sales of ICE vehicles in 2040. Each assumption allows us to calculate the total amount of new vehicles, which are BEV and ICE. Below is a graph that shows the different adoption rates:

With the assumed adoption rates, we see that the percentage of ICE vehicles will reduce over time across our different scenarios, ranging from 97.50% of vehicles being ICE (thus 2.50% being BEVs) in 2050 under the low adoption case, to ~63% of vehicles being ICE (~37% being BEV) under the Rapid adoption+ban of all-new ICE vehicle sales in 2040.

The advantage of calculating our results under different scenarios is that it allows us to see how our outcome variables (energy and emissions) evolve against the baseline scenario, with no introduction of electric cars into the Kingdom. Furthermore, assuming that adoption rates change across time and are not at their highest level under each scenario provides us with a more realistic evolution of the car stock under BEVs and ICE vehicles, as seen above.

Fuel Consumption, estimated kilometres driven, and electric car “fuel efficiency” (Appendix B)

Our second list of variables is regarding fuel consumption, estimated distance driven, and transportation of emissions. First, we obtained the total amount of barrels sold locally from The National Oil & Gas Authority for gasoline and diesel from January 2009 to December 2020 (NOGA, 2020). We then adjust each for seasonal fluctuations via X12-ARIMA. We then total each year of fuel sales so that this allows us to divide yearly total fuel sales by the total amount of vehicles under each year. Our initial equilibrium value is ~18.21 barrels per vehicle, which with our fuel consumption trend, we can forecast till 2050.

Using our monthly seasonally adjusted fuel sales, we then create a linear trend of registered vehicles to estimate the month-by-month total vehicles registered in the Kingdom. This is then used to estimate the fuel consumption per vehicle on a month-to-month basis, where we then calculate the outlier-adjusted average change from 2009 to 2020. As a result, the average month-to-month consumption change is approximate -0.31%, which implies a yearly change of -3.76%.

Using our assumed initial value of 18.21 barrels of fuel sold per vehicle in 2009, we can estimate the fuel consumption per vehicle from 2009 to 2050 with our assumed yearly change. Below is a graph that shows the total litres consumed per year per vehicle:

Our estimated fuel consumption falls from 2900 litres per year to 600 litres. With our fitted fuel consumption trend, we then will estimate the number of kilometres driven. Using IEA data on fuel efficiency for advanced nations with gasoline prices of less than 1 USD (IEA, 2019), in 2005 estimated fuel efficiency average at 11Lge/100km, which declined by 2% per year to 8.6Lge/100km in 2017. Assuming this trend continues, we then calculate our fitted values for fuel-efficiency till 2050 (reducing down to 4.43Lge/100km), using our initial value of 11Lge/100km. Using the estimated fuel consumption per year and our estimated fuel efficiency, we can then calculate the estimated kilometres driven per year (next page on the left).

Page 2

We see that estimated kilometres driven initially start at ~29,000km and fall towards ~14,000km. The graph above reflects only estimates; as to the knowledge of the author, there is no known data that measures the average kilometres driven per vehicle, and this is the first attempted estimate for the Kingdom of Bahrain. The amount of kilometres driven is falling because fuel consumption per vehicle falls faster than the assumed efficiency increases that we have used. This implies that demand for transportation over the long run is falling, given that overall consumption is falling faster than fuel efficiency changes.

Last, we look at our electric vehicles “fuel efficiency.” Using data from the Environmental Protection Agency in the United States (EPA, n.d), we can estimate the “fuel efficiency” of electric vehicles. The data appears to show that electric vehicles average approximately 101.59 Miles per gallon equivalent (MPGE). We then convert this into kWh/100km, allowing us to estimate the total kWh required per vehicle per year.

Transportation Emissions (Appendix C)

For transportation emissions, we calculate the emissions from transportation using L/100km to g CO2/km driven statistics. Specifically, for each 1L/100km of gasoline, this translates to 23.70 g CO2/km; diesel would translate to 26.60 g CO2/km (European Union, 2002). From our seasonally adjusted gasoline and diesel sales data, we fitted a linear trend which shows that the % of diesel sales starts at 36.41% in 2009 and ends at 0% by the year 2050. Given that we do not have data on the number of diesel and gasoline registered vehicles, we create a weighted average g CO2/km figure using the data on the percentage of sales which are diesel and gasoline. Next, using the estimated L/100km from

the previous section and the distance driven, we can calculate the emissions per vehicle in terms of metric tonnes (MT) per vehicle of CO2.

Given that the emissions per vehicle is falling across time, this would mean that our transportation emissions will peak and decline. Below is a graph of the total emissions from transportation in the Kingdom:

We see that Bahrains transportation emission would peak in 2018 at around 3.6 million tonnes, which then declines toward 2.4 million tonnes by 2050 under our assumptions. This fitted plot reflects changes in increased fuel efficiency and the even quicker decline in fuel consumption, which reflects on the distance driven, and, thus, emission per vehicle.

Page 3

Electricity Consumption, Efficiency, and emissions (Appendix D)

Our next set of variables is regarding electricity consumption, efficiency, and emissions. First, we start with consumption. Using data from (IGA Bahrain, 2020) and population data from the world bank (world bank, n.d), we can calculate the consumption per person in kWh/capita. The data shows that the estimated electricity consumption per capita shows a linear trend. Using the linear trend and population forecasts from the World Bank, we then forecast the total gigawatt-hour (GWh ) till 2050.

It is important to note that the graph represents actual and estimated total electricity consumption, where pre-2021 represents actual consumption, and 2021 represents estimated amounts based on World Bank population forecasts and the linear trend on per capita consumption. We see that the total electricity consumption will reach around ~37,000 Gwh in 2050, representing a CAGR of ~2% per year from 2021.

For our electricity production efficiency, we measure the total amount of natural gas per GWh of electricity produced (measured in million cubic feet per GWh produced). This is done by taking data from the National Oil & Gas Authority on the total amount of natural gas provided for electricity production and dividing it by the total amount of electricity produced (NOGA,2020). Once we estimate the amount of natural gas per GWh produced from 2008 to 2020, we then calculate the average yearly change (corresponding to a 7.81% reduction per year on average). Below is a graph representing our fitted natural gas per GWh and the actual measured amount (right side of page).

From the fitted data above, from 2021 to 2050, we calculate the total amount of natural gas used. Here, we assume that production will equate to consumption, meaning that there are no transmission losses or other unaccounted differences between consumption and production. In later revisions of the model, we aim to account for transmission losses and unaccounted differences between consumption and production when developing a more sophisticated model than the initial groundwork.

Last, we look at emissions from electricity production. Using data from the Supreme Council for Environment, we use the total CO2 emissions from energy industries only, and we estimate the total CO2 emissions per GWh produced from electricity production from 2008 to 2015 (SEC, 2020). This is done by taking total energy industries emissions and dividing them by the total GWh production. Then using the average change from 2008 to 2015, we can then estimate future emissions per GWh till 2050, and thus total CO2 emissions. Below is the graph that shows actual and estimated total emissions from electricity production:

Page 4

We can see that our total emissions from energy industries are estimated at 14.75 million tonnes, which then peaks at 15.75 million tonnes and declines towards 14.5 million tonnes of CO2 by 2050. This is because the amount of CO2 per GWh produced reduces significantly across time, from an initial value of 1,012 tonnes of CO2 per GWh to an estimated 401 by 2050.

Results (Appendix E)

Our results will look at energy savings and emissions savings. First, our net energy savings calculations are done by taking our figures of fuel consumption, electricity production, and natural gas inputs for electricity and converting them into terajoules (TJ). We then calculate our net energy savings as a % of overall transportation energy usage and towards Total Final Consumption. This net energy savings reflects the difference between gross energy savings switched towards electric vehicles less the energy required for additional electricity production and natural gas needed for electricity production. Next, we measure our emission savings. Our net emission savings is calculated as the emission savings by switching the car stock from ICE to BEVs and less emissions from additional electricity production. We then calculate this as a net emissions savings as a % of baseline transportation emissions and total national emissions. For this section, we utilize data from the IEA and IGA Bahrain to complete some calculations for the total final consumption of energy and total emissions for Bahrain.

Our final set of results look at how our energy and emissions results would change with the introduction of different renewable electricity production mixes using our rapid adoption case for BEVs.

Net Energy savings as a % of transportation energy usage

We first start with our net energy savings as a percentage of transportation. As previously stated, we calculate our net energy savings as gross energy savings from the reduction of fuel consumption of ICE vehicles that are now BEVs, less additional electricity production required for BEVs, and additional natural gas needed for electricity production. On the right-side of the page are the results.

We see that under the rapid adoption and ban of new ICE vehicle sales by 2040, net energy savings would equate to approximately 17% of our baseline case transportation. This is because the gross savings from switching towards BEVs outweigh the additional amount of electricity and natural gas (for electricity production) used. The results indicate that switching to BEVs would generate significant energy savings for the Kingdom compared to baseline results for transportation.

Net Energy savings as a % of overall total final consumption

Our next results look at net energy savings as a percentage of overall total final consumption. Below are the results:

We see that under rapid adoption of BEV vehicles, the savings equate to nearly 1% of TFC by 2050. This result indicates significant savings as by 2050, under our baseline scenario, transportation would represent 5% of all TFC in the Kingdom with the given assumptions. Overall, it appears that under the rapid adoption case, switching towards BEV vehicles would benefit the Kingdom in terms of curbing energy consumption.

Page 5

Net Emissions savings as a % of baseline transportation emissions

Below are the results of net emissions savings as a percentage of baseline transportation emissions:

In the rapid adoption case, net emission savings would represent 8% of overall transportation emissions, with 37% of the vehicle stock being BEVs. Additionally, under each scenario, as the stock of electric vehicles increases, the savings per year continues to increase, even with the assumption of a 2% increased fuel efficiency for ICE vehicles from year to year. This indicates that under our assumptions, any marginal increase in the percentage of BEVs will lead to a positive marginal increase in the amount of emissions being saved.

Net Emissions savings as a % of baseline overall emissions

Below are the results for net emission savings as a percentage of overall baseline emissions:

We see that overall emissions would decline by 0.51% by 2050 per year. This indicates that switching towards electric vehicles under the rapid adoption and ban of new ICE vehicle sales in 2040 would indicate a substantial emissions savings, given that transportation would represent approximately 6.50% of all emissions by 2050.

different renewable electricity mix

Here we will introduce the different renewable electricity mix strategies that we will introduce into our simplified model. We have derived four different scenarios. Current policies of 10% renewable electricity mix by 2035 (RCREE, n.d), 2x the current aimed policy, 3x the current aimed policy, and 100% renewables by 2040. For the current policy mix, we will assume that the policy will continue to increase the share of renewables to 18.82% by 2050; given that its announcement was made in 2018, the implied increase of shares of renewables would be ~0.59% per year. Below is a graph that shows the differences in the renewable electricity mix for the four different scenarios we introduce:

With all the renewable paths, we assume that each policy would be gradually introduced (in a linear fashion) into the electricity grid. We can then see their effects on emission outcomes with the rapid adoption and the ban of new sales of ICE vehicles in 2040 being the case study we will look at.

Different renewable electricity mix: net emissions savings as a % of transportation emissions

Below are the results of net emission savings as a % of transportation emissions baseline scenario with different renewable electricity mixes.

Page 6

Under 100% renewables mix by 2040, we see that the emissions saved would equate to the gross emission savings. These are simply emissions saved by switching an ICE vehicle toward a BEV vehicle. This is because as the electricity mix reaches a fully decarbonized state (given that renewables do not produce direct emissions), emissions from electricity production become zero. Therefore, having ~37% of the vehicle stock as BEVs would reduce emissions compared to the baseline case by an equivalent amount (~37%).

different renewable electricity mix: net emissions savings as a % of all emissions

Below are the results of net emission savings as a % of all emissions baseline scenario with different renewable electricity mixes.

Under the 100% renewable path, we see that having 37% of the car stock as BEVs would reduce overall yearly emissions by 2.41% by 2050. This is significant given, again, that overall emissions would represent ~6.50%. However, it is essential to note that these results do not take into account the savings of electricity production emissions, where including these figures, by 2050, total yearly emissions would be ~44% lower than the baseline scenario.

Discussions and recommendations

From the results, we see that taking a rapid adoption and banning new sales of ICE vehicles by 2040 would yield favourable results under our simplistic model that we’ve constructed to estimate the effects of energy and emissions outcomes. However, it is important to note that our results would be sensitive to our assumptions. For example, we assumed that fuel consumption would decline by 3.76% per year. However, with our already aggressive assumption that fuel

efficiency would increase by 2% year on year till 2050, this would imply a drastic decline in the total kilometres driven from ~28,500 to ~13,500 in 2050. On the other hand, if we use our figure that does not adjust for outliers, fuel consumption would decline by 2.45%, meaning that our distance driven would only decrease from ~28,500 to ~23,500km. This, in turn, would significantly change our overall baseline figures as transportation emissions would peak by 2033 rather than 2018. Additionally, this would also mean that with the 100% renewables target by 2040, overall emissions savings would nearly double from 2.41% to 4%, with 37% of all vehicles being BEVs.

Despite these limitations, which we discuss in further detail below. The simplistic model shows us three important outcomes. First, Switching towards BEVs would provide a significant potential for energy savings for the Kingdom by 2050. Second, switching towards BEV would also provide significant emissions savings that would allow the Kingdom to decarbonize quicker than the baseline scenario. Last, more rapid adoption of renewables would further magnify the emissions savings by 2050, a critical component given that climate change represents a significant cost to the global economy, which is also significantly higher for neighbouring countries such as Saudi Arabia (Ricke et al. 2018), indicating that Bahrain would also potentially share similar externality costs. Finally, electric vehicles are optimal for at least private transportation needs of the Kingdom because of the relatively small land size, where long driving distances are not a considerable issue. This is confirmed by other studies, such as Bach et al. (2021) white paper, which states that electrification of individual transportation is more appropriate for short to mid-range applications.

The question we should ask ourselves is, how can we bring about the “electrification” of Bahrains transportation system? Furthermore, how do our results compare to research elsewhere? Last, what other considerations can we take from literature outside our simplistic model? To answer the first question, we look at literature which examines subsidies & benefits for BEVs and taxation on ICE vehicles to see their effects on electric car adoption rates.

Although lowering the cost of EV ownership is an important component of adopting BEVs (Kester et al. 2018, Xue et al. 2021, Lavee & Parsha 2021). It is not the most important factor. Other factors, such as infrastructure (primarily accessibility of charging ports), are also incredibly important for EV adoption (Siezchula et al. 2014, Kester et al. 2018, Xue et al. 2021, Lavee & Parsha 2021).

Page 7

Last, there are also other benefits which need to be considered, being campaign awareness on the benefits of electric cars, zoning laws (such as “green zones” or “ultra-low emission zones”), use of bus lanes, and low electricity prices are another crucial determining factor towards the adoption of BEVs (Siebenhofer Ajanovic and Haas 2021, Kester et al. 2018). Furthermore, from a taxation standpoint of ICE vehicles, we see that increasing the cost of ownership for ICE vehicles (via taxation, registration fees, or other policies such as high fuel prices) creates favourable conditions for the adoption of BEVs (Lavee & Parsha 2021, Siebenhofer Ajanovic and Haas 2021).

We then turn to how can we compare our results with research elsewhere. Overall, our results indicate similar benefits in emission reductions, where the overwhelming consensus indicates that switching toward BEVs can significantly reduce CO2, NOx, and PM2.5 emissions (Vilchez et al. 2019, Costa & Seixas 2014, Li et al. 2019). Furthermore, the source of electricity production (the share of clean electricity) also indicates to what extent would emissions be reduced (Li et al. 2019). From an energy savings standpoint to the authors knowledge, there are no papers which specifically look at how the adoption of BEVs would affect overall energy usage. However, despite this, we see that papers have identified that the introduction of BEVs would indeed increase electricity consumption (Kapustin & Grushevenko, 2020) and that the adoption of BEVs in a clustered area would lead to a significant increase in peak load (Muratori, 2018).

From the above, the paper can recommend the following important policies that the Bahraini government should implement to increase the probability of successful rapid adoption of BEVs (right side of page):

The rapid expansion of infrastructure for BEV adoption: A key item needed for the rapid adoption of BEVs is the availability of infrastructure. This means that the need for charging ports and the ability of firms and households to install their charging ports are needed to ensure that charging station availability is high. Furthermore, the need to also ensure that the Kingdom can handle the increased electricity demand from the introduction of BEVs, given that peak load may increase for the Kingdom, given that urban areas are relatively clustered towards the Northern part of the main island and Muharraq island.
Reducing the cost of ownership for BEVs: This would include the use of subsidies, reduction in vehicle registration fees, and other forms of ownership costs that may affect total ownership costs for BEVs. This ensures that the total cost of ownership is at least similar to that of ICE vehicles or lower. A typical citizen and expatriate would look at which method of propulsion is likely to have the lowest total cost of ownership per kilometre driven and therefore choose the cheaper option.
Introduction of additional benefits for BEVs: Additional benefits can vary from the introduction of no parking fees at public parking spots for EVs to the gradual introduction of “green zones” in key areas that may see congestion. This way, a mixture of short-term and gradual introduction of other key benefits may allow owners of ICE vehicles or new car owners to switch towards BEV vehicles.
Increasing the cost of ownership for ICE vehicles: The introduction of higher registration fees, petrol prices, and taxes towards ICE vehicles have proven beneficial for adopting BEV vehicles in conjunction with the other policies mentioned above. These should be conducted gradual and planned, such as the proposed ban of new sales of ICE vehicles in 2040, or a planned increase in petrol prices throughout several years, with welfare support for the lowest income earners who may not have the savings potential to purchase a BEV. This will ensure that their overall welfare is not harmed in the transition process toward an electrified transportation system.

Page 8

Issues and potential remedies for our initial model

Despite the initial model providing some estimates towards energy and emission savings for the Kingdom, several issues can be resolved with additional information to create a more sophisticated model. In the authors opinion, with some assumptions made in this initial model, the results indicated above are potentially underestimated than their actual potential outcomes. The list of issues can be divided into three parts; technical issues with the current model, fundamental issues, and philosophical issues.

Technical Issues

Under technical issues, some assumptions made can be considered strong and unrealistic to some extent. First, we assume that all vehicles are the same and that their fuel economy is based on the IEA's light-duty vehicle fuel efficiency calculations. As of 2020, ~78% of all vehicles in Bahrain are private vehicles that would fit into this description. However, a mixture of other vehicles such as semi-trailers, lorries, buses, ambulances, cranes, and other vehicles differ significantly in terms of fuel efficiency and propulsion systems.

The second technical issue is fuel consumption. As stated above, under the assumptions of a rapidly declining fuel consumption, the total distance driven per year falls dramatically till 2050. This, in turn, would have significant outcomes towards energy usage and emissions outcomes; where in 2021, we estimate transportation represents ~22% of total final consumption of energy, but this rapidly falls towards ~5% by 2050. As for emissions, we estimated that transportation accounts for ~11% of all emissions in 2021 and that this rapidly falls towards ~6.50% of all emissions. Under the counterfactual scenario of using non-outlier adjusted fuel consumption changes, we see that

transportation would represent ~26% of energy usage and ~12% of emissions in 2021, which falls to ~8% and ~11% in 2050, respectively. With the current limitation of data, the next step to solve this issue is to measure actual kilometres driven to derive more accurate driving behaviour and thus a better understanding of the baseline energy and emissions outcomes for transportation.

The third technical issue is that of fuel efficiency increases. We assume that fuel efficiency will constantly increase by 2% each year until 2050. This implies no diminishing effects toward increasing technology growth for ICE vehicles. In reality, the average fuel economy of new light-duty vehicles shows that the average accumulated improvement continues to decline yearly.

If we use the assumptions that fuel efficiency would have diminishing returns towards the rate of improvements, then we would see that by around 2050, the rate of increase for fuel efficiency would be 0.56% This, in turn, would significantly change the results were combined with a more realistic decline in fuel consumption as described in our discussion section would significantly increase energy and emissions savings share.

Page 9

Our third technical issue is our assumed electricity production efficiency. We defined electricity production efficiency as the amount of natural gas consumed per Gwh produced. In 2008, the amount of natural gas used per GWh was 6.79 million cubic feet. This declines drastically towards 0.22 by 2050. It is not entirely clear if such efficiency outcomes are possible. Furthermore, we also assume that production equates to consumption post-2020, another issue that should be considered to be resolved in the future.

Our fourth technical issue relates to the “fuel economy” of electric cars. We assume that electric cars have constant efficiencies, which may not be accurate. Items such as regenerative braking, software improvements, and other key improvements toward electric vehicle characteristics may aid marginal gains in the “fuel economy” of electric cars.

The fifth technical issue is that of electricity production emissions. It is unclear the meaning of “energy industries” CO2 emissions from our data, and the initial value appears to be substantially high compared to other nations. For example, in the United States, natural gas electricity production average emission rates were approximately 407 tonnes of CO2 per GWh in 2017 (EPA, 2018) and 307 tonnes of CO2 per GWh in the United Kingdom in 2020 (Smalldridge, 2021). Furthermore, looking at world data from IEA, we see that if we take the total electricity production from natural gas and simply divide from CO2 emissions from electricity and heat from natural gas energy sources, we see a rough estimate of 492 tonnes of CO2 per Gwh, placing it in line with data from the United States and United Kingdom (IEA, n.d). Our data indicates that in 2019, the estimated emissions per GWh produced is approximately 794, significantly higher than the confirmed and roughly calculated world figure above.

Our last technical issue with the model is in regards to diesel consumption. By 2050, our model assumes there will be no consumption of diesel. This issue ties back to the homogenous car stock issue, where certain vehicles, to the authors knowledge, will always utilize diesel to power the vehicle, thus making this assumption unrealistic.

Ultimately, these technical issues lead to some important conclusions, the model in its current form is likely to underestimate emissions, energy usage, and thus savings from switching ICE vehicles towards BEVs. In reality, it is unlikely that we would see distance-driven decline substantially and that fuel efficiency increases would have diminishing returns, thereby pointing towards a higher contribution of CO2 emissions and energy usage by 2050. Furthermore, other forms of emissions, such as that from power production, is at a much higher initial value, which reduces the net emission savings. Therefore, having a heterogenous car stock, realistic estimates of kilometres driven, more realistic figures of exogenous fuel efficiency improvements, electricity production figures, potential improvements in electric car fuel economy, and emissions from natural gas electricity production will likely yield more realistic emissions and energy assumptions.

Fundamental Issues

A fundamental issue with the model presented above is that it does not necessarily integrate climate and economy outcomes. The model above simply assumes that the Kingdom can change car stock types without any forms of constraints or effects on market outcomes. In reality, changes in market conditions such as changes in production outcomes, real wages, price levels of vehicles, price levels of other goods & services, and other fundamental economic variables would influence the adoption of electric vehicles. For example, what would be the effects of an increase in fuel prices across the Kingdom on the economy (post introduction of subsidized electric cars)? Additionally, what are the effects of changes in income towards distance driven? Creating an integrated assessment model that combines climate and economy would be needed to understand to what extent will the car stock evolve under different policies, what are their economic outcomes and environmental outcomes.

A potential avenue for understanding the effects of electric vehicle introduction on emission and economic outcomes is that of CGE modelling. Under CGE modelling, we can measure and understand the economic and environmental outcomes by introducing policies to encourage the purchase and usage of BEVs. Introducing a policy can allow us to account for interdependencies between different sectors and agents within the economy, thus allowing us to understand their impacts on the wider economy and reveal such policies' indirect or unintended effects.

Page 10

Philosophical issues

A key question should be asked, should we attempt to introduce BEVs to replace the future stock of ICE vehicles, or should we look to conduct transportation differently? Currently, private vehicle usage is the primary mode of transportation used on the island, where we’ve previously established the percentage of mass transportation of overall transportation is low and declining (Rabayah, 2021). Furthermore, we also established that congestion likely has increased given the number of vehicles per kilometre of road available. Given the limited availability of land, are private forms of transportation the most optimal method of transportation?

With the rise of the Bahrain metro aiming to develop 109km of track (Ministry of Transportation and telecommunications, 2022), an optimal policy could be to attempt to substitute consumers behaviours from using private vehicle transportation and instead utilize current and upcoming public transport transportation methods. What would be required is not only public investment and subsidization of public transportation but also changing behaviour and mindsets of the general public towards public transportation, given the current emphasis on using private vehicles as a means of transportation (Abbas, 2018).

By emphasizing the utilization of public transportation over private modes of transportation, energy savings and CO2 emissions (depending on the extent of decarbonization of the electric grid) are likely to be higher than simply switching towards an electrified private transportation system. Private vehicle transportation requires nearly 2.2 times more energy per passenger kilometre (Koe/pkm) than bus and seven times more than rail (Odyssee-Mure, 2021). As a result, even if we see that switching towards BEVs would result in some energy and emission savings, public transportation may be able to provide even more, and thus allow the Kingdom to potentially reduce the environmental burden of energy supply and use, limit costs of energy, or secure it’s energy supply even further.

Conclusion

In this discussion paper, we attempted to model the effects of introducing electric vehicles on energy and emission savings in the Kingdom. We find that if the vehicle stock is 37% electric by 2050 via the rapid adoption and ban of new ICE vehicle sales in 2040, the emission savings and energy savings for transportation and overall economy are significant, especially under the parameters given, which rapidly reduces transportation share of emissions and energy usage. Furthermore, we also show that emissions savings are further magnified by transitioning towards renewables. However, it is important to note that the model is likely underestimating the savings potential and is not as sophisticated as an integrated assessment model of economy and environment.

Last, although the electrification of Bahrains transportation sector would provide significant emission and energy savings (which is likely higher than estimated), an important question which needs to be asked is if private vehicle transportation is the most optimal form of transportation for the Kingdom, and with upcoming newer modes of transportation (Bahrain metro), should we attempt to discourage private transportation for public actively? These questions can be answered in future iterations and the development of our model over several studies toward optimal transportation policies for the Kingdom.

Page 11

Featured

Baland Al Rabayah- Director

Baland Al Rabayah is the co-founder and Director of BRG. He currently holds a Diploma in Accounting and Finance from Bangor University, a Bachelor of Science (BSc) in Accounting and Finance from Bangor University, and a Masters degree in Applied Economics from the University of Strathclyde. Baland will begin his PhD studies at the University of Strathclyde in 2022 looking at the economic effects of renewables on the GCC economies (energy transition). Baland’s interests are in Environmental, Energy, Public Policy, and Health Economics.

email: b.rabayah@bhres.group

SUPPORTING CONTRIBUTOR

*A special thanks to the following member for their contribution towards literature used in this research*

Featured

Mohamed Alamin- Senior Researcher

Mohamed spent over 8 years in the oil and gas industry in various roles. He holds a bachelor degree in mechanical engineering and a master's degree in energy and economics. His interests include energy economics, energy and environmental policy and sustainable development.

email: m.alamin@bahrainresearchgroup.com

Ministry of Information Affairs, M., n.d (Ministry of Information Affairs, n.d). Business and Economy. [online] Ministry of Information Affairs. Available at: https://www.mia.gov.bh/kingdom-of-bahrain/business-and-economy/?lang=en

The Global Carbon Project, T., 2020 (The Global Carbon Project, 2020). Per Capita CO2 Emissions. [online] Our World in Data. Available at: https://ourworldindata.org/grapher/co-emissions-per-capita?tab=chart

EPA, E. n.d (EPA, n.d). All-Electric Vehicles. [online] United States Environmental Projection Agency. Available at: https://www.fueleconomy.gov/feg/evtech.shtml

General Directorate of Traffic, GDT., Ministry of Interior, MOI. , 2020 (GDT& MOI, 2020). Bahrain Open Data Portal. [online] Information & eGovernment Authority. Available at: https://data.gov.bh/

Bahrain News Agency, B., 2021 (bna, 2021). Bahrain’s first electric car charging station inaugurated. [online] Bahrain News Agency. Available at: https://www.bna.bh/en/Bahrainsfirstelectriccarschargingstationinaugurated.aspx?cms=q8FmFJgiscL2fwIzON1%2BDr3U%2BBTvtjBYaFzsn7mXQ3o%3D

National Oil & Gas Authority, N., 2020 (NOGA, 2020). Bahrain Open Data Portal. [online] Information & eGovernment Authority. Available at: https://data.gov.bh/

International Energy Agency, I., 2019 (IEA, 2019). Fuel Economy in Major Car Markets, Technology and Policy Drivers 2005-2017. [online]. International Energy Agency. Available at: https://www.iea.org/reports/fuel-economy-in-major-car-markets

EPA, E. n.d (EPA, n.d). Download Fuel Economy Data. [online] United States Environmental Projection Agency. Available at: https://www.fueleconomy.gov/feg/download.shtml

European Union, E., 2002 (European Union, 2002). Communication from the Commission to the Council and the European Parliament - Implementing the Community Strategy to Reduce CO2 Emissions from Cars: - Fourth annual report on the effectiveness of the strategy (Reporting year 2002) [SEC(2004) 140]. [online] EUR-Lex. Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52004DC0078

IGA Bahrain, I., 2020 (IGA Bahrain, 2020). Bahrain Open Data Portal. [online] Information & eGovernment Authority. Available at: https://data.gov.bh/

World Bank, W., n.d (World Bank, n.d). Population, total-Bahrain. [online]. The World Bank. Available at: https://data.worldbank.org/indicator/SP.POP.TOTL?locations=BH

The Supreme Council for Environmental, T., 2020 (SEC, 2020). Bahrain Open Data Portal. [online] Information & eGovernment Authority. Available at: https://data.gov.bh/

Regional Center for Renewable Energy and Energy Efficiency, R., n.d (RCREEE, n.d). Bahrain. [online] Regional Center for Renewable Energy and Energy Efficiency. Available at: https://www.rcreee.org/member-states/bahrain/4008?page=3

Ricke, K., et al., 2018 (Ricke et al., 2018). Country-Level social cost of carbon. [online] Nature Climate Change. Available at: https://www.nature.com/articles/s41558-018-0282-y#change-history

Bach, C., et al., 2021 (Bach et al., 2021). Pathways to a net zero CO2 Swiss Mobility System. [online] Swiss Competence Center for Energy Research Efficient Technologies and Systems for Mobility. Available at: https://www.sccer-mobility.ch/export/sites/sccer-mobility/aboutus/Network_News/dwn_network_news/210304_WhitePaper2020_RZ_Web.pdf

Kester, J., et al., 2018 (Kester et al 2018). Policy Mechanism to accelerate Electric Vehicle Adoption: A qualitative review from the Nordic region. [online] Renewable and Sustainable Energy Reviews. Available at: https://www.sciencedirect.com/science/article/abs/pii/S136403211830426X

Xue, C., et al., 2021 (Xue et al 2021). Impact of Incentive Policies and Other Socio-Economic Factors on Electric Vehicle Market Share: A Panel Data Analysis from the 20 countries. [online] Sustainability. Available at: https://www.mdpi.com/2071-1050/13/5/2928/htm

Lavee, D., Parsha, A., 2021 (Lavee & Parsha 2021). Cost-benefit analyses of policy tools to encourage the use of Plug-in electric vehicles. [online] Transportation Research Interdisciplinary perspectives. Available at: https://www.sciencedirect.com/science/article/pii/S2590198221001111#b0010

Sierzchula W, et al., 2014 (Sierzchula et al 2014). The influence of financial incentives and other socio-economic factors on electric vehicle adoption. [online] Energy Policy. Available at: https://www.sciencedirect.com/science/article/abs/pii/S0301421514000822

Siebenhofer, M., Ajanovic A., Haas, R., 2021 (Siebenhofer, Ajanovic and Haas, 2021). How Policies Affect the Dissemination of Electric Passenger Cars Worldwide. [online] Energies. Available at: https://www.mdpi.com/1996-1073/14/8/2093/htm

Vilchez, J., et al., 2019 (Vilchez et al 2019). Modelling the impacts of EU countries’ electric car deployment plans on atmospheric emissions and concentrations. [online] European Transport Research Review. Available at: https://www.mdpi.com/1996-1073/14/8/2093/htm

Costa, E., Seixas, J., 2014 (Costa & Seixas 2014). Contribution of Electric Cars to the Mitigation of CO2 Emissions in the City of Sao Paulo. [online] IEEE. Available at: https://ieeexplore.ieee.org/abstract/document/7007035/authors#authors

Li, F., et al., 2019 (Li et al 2019). Regional comparison of electric vehicle adoption and emission reduction effects in China. [online] Resources, Conservation and Recycling. Available at: https://www.sciencedirect.com/science/article/abs/pii/S0921344919300369

Kapustin, N., Grushevenko, D., 2020 (Kapustin & Grushevenko, 2020). Long-term electric vehicles outlook and their potential impact on electric grid. [online] Energy Policy. Available at: https://www.sciencedirect.com/science/article/abs/pii/S0301421519306901

Muratori, M., 2018 (Muratori, 2018). Impact of uncoordinated plug-in electric vehicle charging on residential power demand. [online] Nature Energy. Available at: https://www.nature.com/articles/s41560-017-0074-z

EPA, E., 2018 (EPA, 2018). 2011-2017 Greenhouse Gas Reporting Program Industrial Profile: Petroleum and Natural Gas Systems. [online] United States Environmental Projection Agency. Available at: https://www.epa.gov/sites/default/files/2018-10/documents/subpart_w_2017_industrial_profile.pdf

Smalldridge, G., (Smalldridge, 2021). 2020 UK Greenhouse gas emissions, provisional figures. [online] Department for Business, Energy & Industrial Strategy. Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/972583/2020_Provisional_emissions_statistics_report.pdf

IEA, I., n.d (IEA, n.d). Data and statistics. [online] International Energy Agency. Available at: https://www.iea.org/data-and-statistics/data-browser?country=WORLD&fuel=CO2%20emissions&indicator=CO2EleBySource

Rabayah, B., 2021 (Rabayah, 2021). Should Cars be Taxed? [online] Bahrain Research Group. Available at: https://www.bahrainresearchgroup.com/articles/2021/10/10/should-cars-be-taxed

Ministry of Transportation and Telecommunication, M., (Ministry of transportation and Telecommunication, 2022). Bahrain Metro. [online] Ministry of Transportation and Telecommunication. Available at: https://mtt.gov.bh/content/bahrain-metro

Abbas, H., (Abbas, 2018). How Changes in perception can save the Bahraini bus system. [online] Bahrain Research Group. Available at: https://www.bahrainresearchgroup.com/discussion-papers/2019/1/18/how-changes-in-perception-can-save-the-bahraini-bus-system

Odyssee-Mure, O., (Odyssee-Mure, 2021). Sectorla Profile-Transportation. [online] Odyssee-Mure. Available at: https://www.odyssee-mure.eu/publications/efficiency-by-sector/transport/specific-consumption.html

The appendix is provided as a separate pdf to properly display equations. please click the button below for the appendix.

CLICK HERE