綠色物流專案之淨現值：太陽能板安裝之個案研究 - 政大學術集成

全文

(1)國立政治大學商學院國際經營管理英語碩士學位學程 International MBA Program College of Commerce National Chengchi University. 碩士論文政治. 大. 立Master’s Thesis. ‧. ‧ 國. 學 er. io. sit. y. Nat. 綠色物流專案之淨現值：太陽能板安裝之個案研究 a. n. v Projects: l C for Green Logistics Net Present Value ni hengchi U. A Case Study on Solar Panels Installation. Student: Adrien Pla Advisor: Professor Howard Hao-Chun Chuang. 中華民國106年7月 July 2017.

(2) 綠色物流專案之淨現值：太陽能板安裝之個案研究 Net Present Value for Green Logistics Projects: A Case Study on Solar Panels Installation 研究生：林文龍. Student: Adrien Pla. 指導教授：莊皓鈞. 立. Advisor: Howard Hao-Chun Chuang. 政治大. ‧ 國. 學. 國立政治大學. 商學院國際經營管理英語碩士學位學程. io. y. sit. Nat. A Thesis. ‧. 碩士論文. a. er. Submitted to International MBA Program. n. iv National Chengchi University l. n U e n gofcthe h iRequirements in partial fulfillment. Ch. for the degree of Master in Business Administration. 中華民國106年7月 July 2017.

(3) Acknowledgements. I would like to express my gratitude to Professor Howard Chuang, my thesis advisor, for his great support and extensive knowledge to go through the barriers encountered during my work. I thank the jury present during my oral defense for taking their time to evaluate my work. I would also thank NCCU, for giving me the chance to achieve this academic study, and. 政治大. particularly the IMBA office to have help me through all the process.. 立. I also would like to thank my parents for their moral and financial support, my friends, as well. ‧ 國. 學. as the Region Haut-de-France for granting me a scholarship this semester.. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. i. i n U. v.

(4) Abstract Net Present Value for Green Logistics Projects: A Case Study on Solar Panels Installation By Adrien Pla The gravity of the Global Warming consequences force us to face the issue and try the best to. 政治大. mitigate its effects. Green Logistics practices, and especially the case of a Solar Panels. 立. Installation on a warehouse or plant’s roof, are studied in this article as solutions to reduce the. ‧ 國. 學. economic, social and environmental casualties. The Net Present Value (NPV) Model in this paper and its results analysis show that project duration, Carbon Tax, initial energy price and. ‧. its increasing rate are all positively related to the NPV Value. A high discount rate will,. Nat. sit. n. al. er. io. projects.. y. however, turn the NPV negative, thus raising the need of financing help for Green Logistics. Ch. i n U. v. Keywords: Net Present Value, Green Logistics, Carbon Tax, Solar Panels Installation, Global Warming. engchi. ii.

(5) TABLE OF CONTENTS. 1. Introduction .......................................................................................................................... 1. 2. Literature review .................................................................................................................. 7. 2.1. Net Present Value ............................................................................................................. 7. 政治大. 2.2. Green Logistics ................................................................................................................ 8. 立. 3. CO2 equivalent Emission and Carbon Tax ...................................................................... 11. ‧ 國. 學. 4. A NPV model for Solar Panels installation ....................................................................... 17. ‧. io. sit. y. Nat. 4.1. Installation Cost ............................................................................................................ 19. er. 4.2. Cash Flow ...................................................................................................................... 21. al. n. v i n Ch 4.3. Rate of return ................................................................................................................. 23 engchi U 5. Analysis & Results .............................................................................................................. 24. 6. Discussion & Conclusion .................................................................................................... 32. Reference ................................................................................................................................. 35. Appendix ................................................................................................................................. 38. iii.

(6) List of Figures and Tables Figure 1: Greenhouse gases and their Global Warming Potential ....................................... 2 Figure 2: Solar Panels on the Logistic base of Galbani, Group Lactalis Italy .................... 4 Figure 3: Installed Price, Module Price Indec, and Implied Non-Module Costs over time. 政治大. for Residential PV systems ............................................................................................... 5. 立. Figure 4: Sustainable Supply Chain Management overview ................................................ 9. ‧ 國. 學. Figure 5: Generation by Fuel Type and CO2 ....................................................................... 14. ‧. Figure 6: Impact of the Increase Rate and Carbon Tax ……………………….…………. 26. Nat. io. sit. y. Figure 7: Impact of Initial Price of Energy and Carbon Tax ……………………………. 28. n. al. er. Figure 8: Impact of Length of Project and Carbon Tax …………………………………. 30. Ch. engchi. i n U. v. Table 1: The NPV model elements ………………………………………..………….... 18-19 Table 2: Base Values of Variables & Parameters ………..……………………………….. 25. Appendix: Quotation - 10kWh Energy Storage System with Solar Panels ……...…….. 38. iv.

(7) 1. Introduction. The global warming with the greenhouse effect is described as the problem of the century: “Climate change is the biggest externality our planet might face.” (Rezai et al. 2012). Its consequences are an increasing volatility of natural disasters, a polarisation of climate effect and a raise of the level of water reducing habitable land. A problem which is significant when. 政治大 the sea level (Zhang, Douglas and Leatherman, 2004). Another consequence of global warming 立. coastal areas are attractive for mankind, indeed 100 million people living within one meter from. ‧ 國. 學. is the increasing aridity in low precipitation area. Severe draughts’ impact on crops yields in certain region will be too negative to be countered by improving agricultural process (Cook et. ‧. al., 2014). Effects on ecology will also be dramatic “A large fraction of species faces increased. sit. y. Nat. extinction risk due to climate change during and beyond the 21st century”. (IPCC, 2015).. io. al. er. Consequences of global warming will be too violent and significant to be countered by mere. v. n. reaction and adaptation. A proactive approach is required to effectively mitigate the risks.. Ch. engchi. i n U. However, the reasons of global warming are known. Greenhouse effect is caused by the emission of different gases. Carbon Dioxide (CO2), which is responsible at a level of 76%, Methane (CH4), 16%, Nitrous Oxide (N2O), 6% and Fluorinated gases (F-gases), 2%. (IPPC, 2015). Carbon Dioxide equivalent or CO2e regroups the Greenhouse gases to a single unit. “For any quantity and type of greenhouse gas, CO2e signifies the amount of CO2 which would have the equivalent global warming impact.” (Brander, 2012). The global warming impact is not the same for the different greenhouse gases. The global warming potential of greenhouse gases takes as index the increase of temperature caused by CO2 emission over 100 years.. 1.

(8) Figure 1: Greenhouse gases and their Global Warming Potential (Adapted from IPCC, 2007). As observed in Figure 1, an emission of 1 kg of Methane will be computed as 25 kg of carbon. 政治大 impact on the next 100 years than the one of 1 kg of CO2. The other environmental issue where 立 dioxide equivalent (CO2e) because an emission of 1 KG methane will have 25 more warming. ‧ 國. 學. results are more immediate is air pollution, which is a local issue caused by other gases. Its consequences are disastrous, “Polluted air was responsible in 2015 for 6.4 million deaths. ‧. worldwide” the biggest factor after tobacco which was responsible of 7 million deaths.. sit. y. Nat. (Landrigan, 2017). Pollution through logistics affects more global warming with the. io. er. greenhouse effect than air pollution, as distribution centers, plants and most of goods. al. v i n industrial center is one of the C reason h efor i U reclusive position. To counter the n gtheir c hrelative n. transportation and transformation are done in remote area. Pollution brought by factories and. pollution brought by logistics operations, new practices such as Green Logistics, in relation with the global warming issue awareness, emerge.. Green Logistics as defined by Dekker et al. (2011), “is the study of practices that aim to reduce the environmental externalities, mainly related to greenhouse gas emissions, noise and accidents, of logistics operations”. As the issue of global warming is very large, a unique solution to mitigate its risks does not exist, there are many. A specific approach must be taken to tackle the problem little by little. The idea of the project is to create a useful tool for logistics. 2.

(9) project implementation in the business field. This tool using the Net Present Value (NPV) will give a financial evaluation and a CO2e emission measure of the project. This measure will be in kg emitted and in dollars. While NPV measures the economic sustainability of a project, an environmental layer will be added. The environmental aspect is measurable and can be priced mixed with the economic one. Still, showing the emission of pollutant will be an environmental criterion in decision making. A weighted decision criteria can also be reviewed to help people choose the best project for the company and for curbing the global warming.. 政治大. 立. While there are many different possible Green Logistics projects, for illustrations purposes,. ‧ 國. 學. only the solar panels installation on warehouses’ or plants’ roof will be considered in the cost model. The choice of a solar panels installation was motivated by the following reasons.. ‧. Installing solar panels on a warehouse is a simple concept and a talkative example. Moreover, it. y. Nat. sit. is feasible economically and technically. Figure 2 shows a prototypal example of how solar. n. al. er. io. panels installation has been successfully implemented on a national refrigerated distribution. i n U. v. center for an international dairy company in Ospedaletto Lodigiano, Italy. By doing so, the. Ch. engchi. distribution center reduces its environmental impact by reducing its emissions of CO2e and may reduce its expenses on energy.. 3.

(10) Figure 2: Solar Panels on the Logistic base of Galbani, Group Lactalis Italy. 立. 政治大. ‧. ‧ 國. 學 er. io. sit. y. Nat. Nevertheless, implementing solar panels as energy supply has a cost, even though the cost is. n. al. Ch. i n U. v. decreasing. The Figure 3 shows its decreasing cost trend from 1998 to 2015 for residential. engchi. photovoltaic (PV) systems. In addition to decreased costs over time, the benefits from the installation are the savings on the energy bill thanks to the supply of energy by the company itself. In addition, solar panels offer an energy almost free of CO2e so the company will be able to avoid a certain amount of Carbon Tax, if one is implemented in the country the firm is operating in.. 4.

(11) Figure 3: Installed Price, Module Price Indec, and Implied Non-Module Costs over time for Residential PV systems (Adapted from Barbose and Dargouth, 2016). 立. 政治大. ‧. ‧ 國. 學 sit. y. Nat. io. er. In response to the emergence of Green Logistics initiatives, and particularly the practical need. al. v i n C hof the price of energy, been developed: What is the impact e n g c h i U its variation, project duration, and n. for assessing cost-efficiency of solar panels installation, the following research questions have. the level of Carbon Tax on the NPV of solar panels installation? Under what conditions will the installation be economically viable?. To answer the research questions, the study consists of a literature review in Section 2 and a in depth discussion of Carbon Tax in Section 3. A cost model that has been developed is presented in Section 4. Created in a professional environment in the first place, the model has been rearranged, modified, and enhanced to fit research purposes. After obtaining estimates of the NPV Model variables and parameters, a comprehensive analysis will be performed to answer. 5.

(12) the research questions in Section 5. Finally, the results, limitations and contributions will be discussed in Section 6.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 6. i n U. v.

(13) 2. Literature review. The literature review will cover two important aspects. The first one, financial, is the net present value method. Its definition, application and its relevance in this article will be covered. The second aspect, operational, is Green Logistics.. 2.1. Net Present Value. 立. 政治大. ‧ 國. 學. The NPV “returns the net present value of the investment based on a series of periodic cash flows and a discounted rate.” “NPV is derived by discounting future net cash flows” by the. ‧. discount rate r, “summing them over the life of investment, and deducting the initial investment. sit. y. Nat. related to the proposal”. (Lin and Nagalingam, 2000). NPV is a commonly used technique for. io. al. er. evaluating investment options and projects. If the NPV is positive, then the project is. n. economically acceptable. This condition is compulsory for most of the business decisions for a. Ch. company to be sustainable financially.. engchi. i n U. v. Even though NPV is easy to use and relevant, it has potential limitations. NPV Value is qualified of “suboptimal investment decision” under two conditions, if sunk costs are included and time length of the project is not set and too flexible (Doraszelski, 2001). The value option of waiting is not considered in the NPV but investment timing is important, a sub optimally adopted project can lose from 10 to 20 percent of its value (McDonald and Siegel, 1986). However, the option of waiting has also costs. Firstly, waiting delays the receiving of cash flow generated by the project. Secondly, in the case of a Green Logistics project, the cost of waiting is beard by the environment because CO2 emissions reductions are also postponed. The level of 7.

(14) criticality of the global warming issue makes the option value of waiting for implement green projects less and less affordable. The effect of the rising oceans induced by a non-mitigated global warming would be beyond measure and massively negative. “The economic and social cost of losing functionality of all coastal cities is practically incalculable.” (Hansen et al., 2016).. The NPV is also used for the sake of its simplicity. More than its quantitative correctness, the NPV found its value in the qualitative assumptions, “in the face of extreme uncertainty. 政治大. quantitative analysis has difficulty to outperform qualitative analysis” (van den Bergh, 2004).. 立. This paper will use both sides to make the cost model a helping tool for decision making in term. ‧ 國. 學. of Green Logistics projects.. ‧. n. al. er. io. sit. y. Nat. 2.2. Green Logistics. i n U. v. The Green Logistics practices aiming at reducing environmental externalities of logistics. Ch. engchi. operations (Dekker et al., 2011) fit in a bigger concept: Sustainable Supply Chain Management (SSCM). SSCM defined by Carter and Rogers (2008) is “the strategic, transparent integration and achievement of an organization’s social, environmental, and economic goals in the systemic coordination of key interorganizational business processes for improving the long-term economic performance of the individual company and its supply chains.” (Carter and Rogers, 2008). The Figure 4 shows an overview of SSCM where the concept of the triple bottom line, first cited by John Elkington in 1994, is the base. People, planet and profit are the three Ps for social, environmental and economic performance (The Economist, 2017).. 8.

(15) Figure 4: Sustainable Supply Chain Management overview (Adapted from Carter and Rogers, 2008). 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. i n U. v. The Figure 4 suggests that a project performing only on the environmental and social aspects. Ch. engchi. can be a good project. However, a positive economic return is compulsory for a project to be sustainable. Unless it is an only a side project or if it is implemented by a non-business entity, such as Non-Governmental Organization or a government. In addition to installation of solar panels on warehouses or plant’s roofs, there are various Green Logistics projects. For example, the replacement by electric vehicles (EV) of former fleet can be effectuated. An upfront investment must be done to replace combustion engine vehicles by electric ones but the cost can be offset by the lower cost of carburant and the better efficiency of EV and a reduction of CO2e emission and air polluting gases. Green packaging design and. 9.

(16) implementation is also an example of Green Logistics project. Using smaller volume requires less transportation to move the same number of goods and designing packaging material to be recycled or re-used diminishes the amount of material needed, so the pollution induced. Finally, carbon offsetting projects should be mentioned. Their aim is to take away the CO2 from the atmosphere while the previous project only reduces the emissions. The carbon offsetting projects such as planting trees allows some entities to pursue a zero-emission objective such the one proposed by Elioth Consulting presented to the city of Paris, “Paris change d’ère” (Le Monde.fr, 2017).. 立. 政治大. ‧ 國. 學. Another objective of Green Logistics project is the positive image return to the company, the effort toward mitigating the global warming issue is more and more valorized by customers.. ‧. “Besides attracting new profitable customers for organizations, it will give a competitive edge. sit. y. Nat. in the market place. It will also strengthen the brand image and reputation in the market place.”. io. al. er. (Emmett and Sood, 2010). Nevertheless, the positive image effect of Green Logistics effects. n. will not be considered in the quantitative analysis as consequences, mostly positive, are hard to quantify and estimate.. Ch. engchi. 10. i n U. v.

(17) 3. CO2 equivalent Emission and Carbon Tax. Green Logistics projects reviewed in 2.2, such as the solar panels installation on a warehouse considered in the study, have a positive environmental impact. They reduce the emission of CO2e in the atmosphere slowing the global warming. To further mitigate the global warming issue, a financial cost for CO2e must be set. A carbon price will reward companies doing effort. 政治大 are several ways to price carbon dioxide emissions such as carbon tax, cap and trade, carbon 立. to reduce their emission while punishing the ones neglecting their environmental impact. There. ‧ 國. 學. trade, and carbon credit. The carbon tax is simpler and more easily integrated in investment models compared to the other types of carbon pricing. Being one of the main components of the. ‧. cash flow in the model being developed, the carbon tax is an important aspect to be analyzed. Its. sit. y. Nat. goal is to “drive emissions reduction and mitigate the risks posed by climate change” (Aldy,. io. er. 2017). Implementation of carbon tax has increased a lot in the last ten years and is expected to. al. grow in the near future. “2017 could see the largest ever increase in the share of global. n. v i n C hinitiatives in a single emissions covered by carbon pricing e n g c h i U year.” (World Bank, 2016).. Aldy (2017) proposes a model where the price of carbon tax should be alienated with its need to counter the global warming effects. If the consequences of the climate issue are worse than expected, then the carbon tax price will rise. If better than expected, then the carbon tax will reduce. Of course, evaluating consequence of such a global and massive problem can be troublesome but it would be based on scientific research. The trend of new publications and scientific discoveries will orientate the carbon tax. “Some uncertainties about climate changes will be resolved” over time, their conclusions should be included in the carbon tax (Aldy, 2017).. 11.

(18) However, changes in the scientific field are continuous and the price would change all the time. And “The lack of predictability in the carbon price over time could undermine business and household planning and investment.” (Aldy 2017). The NPV model estimates investment value on project having till twenty years’ time length, sometimes more. The Carbon Tax should be predictable to be included on the model and help to the decision making. A higher certainty will need less risk compensation for a project to be validated and therefore increase its value.. 政治大. The carbon price is related to three types of uncertainties regarding environment, international. 立. relations and economics. Environmental uncertainty is related to the estimation of risks due to. ‧ 國. 學. global warming. If the risks are under evaluated than the carbon tax will be increased and if risks are overestimated, then it will be decreased. The politics uncertainty relates to the decision. ‧. time length between necessary changes that can be too long compared with the need to make it.. y. Nat. n. al. er. io. sit. Therefore, updates are not synchronized with environment conditions.. i n U. v. As the global warming issue affects all countries, it is an international problem that has to be. Ch. engchi. mitigated nationally with comparable effort. Recent agreements are based on initiative over compliance. The condition for success is if first steps are taken and if there is a mutual observance between countries, bringing a virtuous circle of responses toward the global warming issue. There is a need a certain transparency between countries which is currently far to be sufficient (Aldy, 2017), a system of reviews from each country with comparable criteria. Carbon tax and other carbon pricing methods were put into the latest and most important agreement: “The Paris Agreement lays the basis for facilitating international recognition of cross-border approaches to cooperation on emissions mitigation, including pursuing cooperation through an international carbon market.” (World Bank, 2016). 12.

(19) To reach the objective of Paris agreement of limiting the increase in temperature of 2 degree Celsius the price of carbon was estimated to “be between US$80/tCO2e to US$120/tCO2e in 2030” (World Bank, 2016). More and more countries set up a carbon tax to fight against global warming. Sweden is currently pricing the highest with $131 per ton of CO2e emitted (World Bank, 2016). When a company implements solar panel installation, the energy got is almost CO2e free, except from CO2 emitted during fabrication and a transformation of components,. 政治大. which is not considered in the model. Solar energy use is CO2 emission free whereas grid. 立. energy (regarding the type of energy plants) or diesel generator emits lots of carbon dioxide and. ‧ 國. 學. equivalents in the atmosphere. With a carbon tax implemented, not emitting CO2 has both environmental and economic value with less gases contributing to the global warming and. ‧. incremental savings for the company.. n. er. io. sit. y. Nat. al. Ch. engchi. 13. i n U. v.

(20) Figure 5: Generation by Fuel Type and CO2 (Adapted from US EPA, 2017). 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. The number of kg of CO2 emitted by the average energy company per kWh is a local data. It depends the country or even the region the energy grid is located. For example, the Environmental Protection Agency in the USA computed the emissions of greenhouse gases of energy production for the year 2014 for each state. As seen in the Figure 5 above CO2 emission rates strongly vary between states because of their different energy mix. However, in the USA, in 2014, the overall average non-based loaded emitted CO2e is 0.704 kg per kWh of energy (1552.5lbs CO2e/ MWh). For a matter of simplification this number will be used in the simulations to compute the Carbon Tax Avoided in section 5. “Annual non-baseload output. 14.

(21) emission rates […] can be used to estimate greenhouse gases emissions reductions from reductions in electricity use.” (US EPA, 2017). In our model using energy harvesting from self-implemented solar panels means reducing the energy consumption of the company from the grid therefore the non-baseload emission rate will be used. The baseload emissions are “the emissions from power plants that run all the time” non-baseload generation are from “power plants that are brought online as necessary to meet demand.” (US EPA, 2017). 政治大. Moreover, predictability of Carbon tax is crucial, the challenge is to find the right balance. 立. between predictability and correctness. A lack of predictability has an adverse effect on. ‧ 國. 學. investment planning. The Carbon price should be “know for many years into the future” and be corrected regarding environment and international conditions. Stating its variables and. ‧. explaining them could increase the predictability of the Carbon Tax. An education program to. y. Nat. sit. make people sensible to the tax carbon as a measure to counter global warming would increase. n. al. er. io. its acceptance an implementation.. Ch. engchi. i n U. v. A higher price of energy which is a consequence of Carbon tax is also an argument against it. It would reduce purchasing power of citizens and harm the whole industry. However, the impact of it can be relativized. Oil being the main world energy source with 44% in 1971 and 31% in 2014 (International Energy Agency, 2016), oil price sharp increase have a smaller impact than expected. “A 10 percent innovation in the price of oil is predicted to contract private sector output by about one-half a percent” (Rotemberg and Woodford, 1996). A tax carbon would be bearable for the economy. The tax collected could be used to help technical innovation toward reducing CO2 and equivalents emissions. “It is more important to have substantial R&D subsidies for green technology to kick - start green innovation and fight global warming.” 15.

(22) (Rezai et al. 2012). Following Aldy’s logic, if innovations mitigates more than expected the risks of global warming then the carbon tax can be reduced, or show a slower increase. The better global warming risks are mitigated, the less needed a carbon tax is.. After estimating its relevance in literature, the carbon tax will be included in the model in Section 4 and its financial impact will be assessed in Section 5.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 16. i n U. v.

(23) 4. A NPV model for Solar Panels installation. The following example of Green Logistics project is the installation of solar panels on warehouses’ or plants’ roof. Using the NPV cost model considers the project as an investment. The cost model can be useful to the investor who will decide to implement the project and to the seller of the solar panel systems. Having a solid cost model will allow the seller to establish a. 政治大 mutual understanding will allow easier negotiations and deals will be profitable on both sides 立. fair price for its products and installation. If estimations on both side are close to each other the. ‧ 國. 學. and sustainable. The technique shows the solar panels installation from a rational investor point of view. It aims to be transparent on all form of costs and cash flows.. ‧ sit. y. Nat. The equation 1 is the main one of the NPV model for the Green Logistics project where t is the. io. n. al. er. year index, CFt denotes annual cash flow, and r the discount rate.. Ch. engchi. iv n U 𝑇. 𝑁𝑃𝑉𝑆𝑜𝑙𝑎𝑟 = − 𝐼𝑛𝑠𝑡𝑎𝑙𝑙𝑎𝑡𝑖𝑜𝑛 𝐶𝑜𝑠𝑡 + ∑ 𝑡=1. 𝐶𝐹𝑡 (1 + 𝑟)𝑡 (1). The NPV model for Solar Panels installation is composed of several equations with multiple variables. The Table 1 defines and describes all the variables and parameters of the model.. 17.

(24) Table 1: The NPV model elements Variables & Parameters. Unit. Description. Installation Cost. USD ($). Fixed. costs. of. the. project. including. design,. components and setting costs. Generally paid upfront (at t0). (Eq 1) CFt. USD ($). Annual cash flow generated by the project for the year t. (Eq 1). 治rate of the project including the risk inherent 政Discount 大. r. to it and the real interest rate. (Eq 1). USD ($). Pt is the Energy Price the company should pay to an. ‧ 國. 立 sionless. energy suppliers at year t for one kilowatt-hour. It is. ‧. sit. kWh. Energy collected through solar panels at year t. (Eq 2). er. io. aUSD v l C($) Yearly amount ofn i tax hengchi U. n. Carbon Tax Avoidedt. equation 3). (Eq 2). y. compounded with an increase rate i (following. Nat. Solar Panels Output t. 學. Pt. Dimen -. related to CO2e emission. avoided thanks to the project. (Eq 2). Maintenance Cost t. USD ($). Yearly running cost including verification, cleaning and replacement of components such as inverter or Maximum Power Point Tracking devices (MPPT).. Daily Marketed Output. kWh. Energy collected in average by the new solar panels in one day.. Price of Carbon Taxt. USD ($). Determined by the government, it is a tax on the kilogram of CO2e emitted by companies. (Eq 5). 18.

(25) CO2e per kWh. Kg. Kg of CO2e emitted to produce one kWh of energy (Eq 5). Po. USD ($). Initial price of energy, chosen regarding the location of the project and the type of energy supply.. Decay Ratet. Dimen -. Mathematically represents the loss in efficiency of a. sionless. solar panel through its utilization life.. T. Years. Count of t, it is the total length of the project life. i. Dimen -. 政Increase治rate 大 of energy. 立 sionless. price, chosen regarding. different scenarios, it is a constant value.. ‧ 國. 學. The following sections introduce equations 2 to 6 that are all linked to the equation 1. Equation. ‧. 1 has 3 variables. The first one, the installation cost is the upfront cost of the system. The second. y. Nat. sit. variable is the cash flow (CF) generated by the savings which is discounted over the years by r,. n. al. er. io. the discount rate and the last variable. It is used to represent the time value of money and risk.. 4.1. Installation Cost. Ch. engchi. i n U. v. The Installation Cost is the upfront cost of the product including the cost of components used in the making of the system and the set-up cost. The cost of components is the product sum of the number of each component needed with its unit cost. The assembly of those components is counted in the price of component as well. The installation cost includes the number of person-day needed for the installation to be complete time the cost of one person-day including the wage. A person-day is the productivity measure of what a person can do in one day. For. 19.

(26) example, 10 person-days could be 1 person working during 10 days, or 10 persons for 1 day, 5 persons for 2 days, and so on. The product shipping cost and the personnel expenses such as plane tickets, accommodation and food are also included in the Installation Cost.. Investment wise the Installation Cost is a negative value which is paid right away, so there is no need to discount it. The Installation Cost is in function of the size of the project therefore in the example of the surface of solar panels. The Installation Cost can be related to the Daily. 政治大. Marketed Output by a ratio of dollars per kilowatt-hour.. 立. ‧ 國. 學. The different components are the solar panels, the racking to maintain the panels to get the best output possible and the cables to conduct the energy. The batteries are here to store the energy. ‧. generated by solar panels. In the same way as the solar panel output, the capacity of batteries is. y. Nat. sit. slowly reducing over time. Nevertheless, the battery capacity is not considered in Equation 2. n. al. er. io. because it only stores energy, does not create some on its own. It is however compulsory in the. i n U. v. process. The reason is to counter the volatility of solar panels output, mostly for weather. Ch. engchi. conditions. Stocked in the batteries in case of surplus, the energy then can be used when requested and does not have to be consumed right away. Thus, the cost of batteries needs to be included in the Installation Cost. Another component to describe is the MPPT (Maximum Power Point Tracking) that optimizes the energy generated from solar panel to the batteries. Inverters are present to convert the direct current to alternative current. Direct current or DC is the type of voltage or current in which the energy is harvested from the solar panels. The second type of current, alternative or AC is used by most of our devices as it is the most convenient way to transport energy on long distances. A system with inverters can supply energy to devices requiring DC and/or AC. 20.

(27) 4.2. Cash Flow. The cash flow is the annual savings achieved thanks to the installation of the system. It is represented by the following two equations: 𝐶𝐹𝑡 = 𝑃𝑡 × 𝑆𝑜𝑙𝑎𝑟 𝑃𝑎𝑛𝑒𝑙𝑠 𝑂𝑢𝑡𝑝𝑢𝑡𝑡 + 𝐶𝑎𝑟𝑏𝑜𝑛 𝑇𝑎𝑥 𝐴𝑣𝑜𝑖𝑑𝑒𝑑𝑡 − 𝑀𝑎𝑖𝑛𝑡𝑒𝑛𝑎𝑛𝑐𝑒 𝐶𝑜𝑠𝑡𝑡. (2). 𝑃𝑡 = 𝑃𝑜 (1 + 𝑖)𝑡 (3). 政治大. Energy Price (Pt) is what the company currently pays for one kilowatt-hour at year t. It is. 立. measured in US dollars. The energy has a price which can be expected to variate along the. ‧ 國. 學. project life. Future variations can be integrated in the model with a constant increase (i) over the years for example. The price per kWh times the yearly energy output from panels (in kWh) is. ‧. the amount of money saved on energy bills thanks to the system. Therefore, the higher the price. y. Nat. n. al. er. io. sit. of energy the more the system is profitable.. 𝑆𝑜𝑙𝑎𝑟 𝑃𝑎𝑛𝑒𝑙𝑠 𝑂𝑢𝑡𝑝𝑢𝑡𝑡. Ch. engchi. i n U. v. = 𝐷𝑎𝑖𝑙𝑦 𝑀𝑎𝑟𝑘𝑒𝑡𝑒𝑑 𝑂𝑢𝑡𝑝𝑢𝑡 × [1 − 𝐷𝑒𝑐𝑎𝑦 𝑅𝑎𝑡𝑒 × (𝑡 − 1)] × 360 (4). Solar Panels Outputt is the energy collected from the sun, it will replace the one bought from the grid or any other type of generator. It is measured in kilowatt-hour at year t. The energy that would have been bought without the system is calculated by the solar panels output. A solar panel has a marketed daily output which slowly decrease as time goes. Solar panels have warranty of minimum output for a determined period. The depreciation effect is represented by a rate of decay. For example, if a warranty of 80% efficiency after a period of 10 years of 21.

(28) utilization is included, it can be considered that the efficiency will be diminished by 2% per year. The Decay Ratet times the marketed output is the real daily output. Multiplied by 360, assuming 360 days in a year, the equation (4) gives the yearly output of solar panels.. 𝐶𝑎𝑟𝑏𝑜𝑛 𝑇𝑎𝑥 𝐴𝑣𝑜𝑖𝑑𝑒𝑑𝑡 = 𝑃𝑟𝑖𝑐𝑒 𝑜𝑓 𝐶𝑎𝑟𝑏𝑜𝑛 𝑇𝑎𝑥𝑡 × 𝐶𝑂2𝑒 𝑝𝑒𝑟 𝑘𝑊ℎ × 𝑆𝑜𝑙𝑎𝑟 𝑃𝑎𝑛𝑒𝑙 𝑂𝑢𝑡𝑝𝑢𝑡𝑡. (5). 政治大. The Carbon Tax avoided is composed of the Carbon tax per kilogram of CO2 emitted times the. 立. number of kilograms of CO2e emitted by average energy company per kWh times the yearly. ‧ 國. 學. energy output of the system. The Carbon Tax per kilogram of CO2e or Price of Carbon Tax is decided by the government of the country the company is operating in. Its amount as discussed. ‧. later can vary regarding different scenarios. As written in section 3, to reach the objective of the. y. Nat. n. al. er. io. sit. Paris agreement the Carbon price should be between 80$ and 120$ (World Bank, 2016).. i n U. v. The last factor of the equation (2) CFt is the Maintenance Cost. It includes the cost of. Ch. engchi. components needed to be replaced and the personnel cost for the maintenance activities per year such as cleaning solar panels, checking wires, verifying the charging of the batteries and so on.. The equation (2) gives the annual cash flow provided by the system. Nevertheless, as CFt is future savings, it needs to be discounted over a certain period. This period T is usually determined by the utility life of the project, by how long it will and how long it can be used. The time length should fit the specificity of the project in question. Usually a longer period will show better results as the financial weight of fixed cost decreases over time, countered by the cash flows. However, high maintenance cost could make the previous statement false. The 22.

(29) warranty on appliances can be a good indicator of the lifespan of the project. In the case of solar panels installation. If panels are warranted 10 years to give at least 80% of the marketed output, it is safe to calculate that the real output of appliances will be the same as the one guaranteed.. 4.3. Rate of return. The discount rate r, in our NPV model (equation (1)), is composed of the risk-free rate and the. 學. 𝑟 = 𝑟𝑖𝑠𝑘 𝑓𝑟𝑒𝑒 𝑟𝑎𝑡𝑒 + 𝑟𝑖𝑠𝑘 𝑝𝑟𝑒𝑚𝑖𝑢𝑚 (6). ‧. ‧ 國. 治政 risk premium related to the project. 大立. The discount rate r has two roles. The first is to represent the monetary value of time, which. y. Nat. er. io. sit. include the phenomenon of inflation. A certain amount of money worth more now than the same amount in the future. The second role is to represent the level of uncertainty associated to. n. al. Ch. i n U. v. the investment. A high uncertainty implies a high return as compensation. In a way, it is also the. engchi. opportunity cost of the project. The amount invested could have be used to buy a safe bond with an assured return on investment. The risk-free rate in the equation represent the last assumption. The 10 year US bond rate is commonly used for risk free rate. As many factors, hard to precisely estimate, enter in the composition of r the best to be done is an educated guess. However, its presence is compulsory to link the NPV to a realistic amount. As different discount rates represent different level of risks and different approach of investment r can be chosen as parameter.. 23.

(30) 5. Analysis & Results. The case used for the model and the analysis is an installation of solar panels on a warehouse or plant’s roof. The objective of investing in such a project for a company is double. The first part is for the company to increase free cash flow like every other business projects. The second part is environmental by reducing the emission of carbon dioxide and equivalents in the atmosphere,. 政治大. thus reducing the global warming effect.. 立. ‧ 國. 學. The green logistic project to install solar panels on a warehouse or plant’s house is based on many different assumptions. A sensitivity analysis gives a comprehensive view of the different. ‧. outcomes regarding the project. The NPV is the dependent variable which will be affected by. sit. y. Nat. the independent variables or parameters. After the analysis, relations between dependent and. io. er. independent variables will be determined and their extent as well. The data used during the. al. v i n Gloria C Energy Limited, Tesla h e nAsset gchi U. n. sensitivity analysis, presented in the Table 2 below, comes from the energy professional environment including. Environmental Protection Agency.. 24. Energy Storage, and the.

(31) Table 2: Base Values of Variables & Parameters Variables & Parameters. Base Value. Installation Cost. $ 12548. Length of Project T. 20. Maintenance Costt. $ 200. Initial Price of Energy P0. $ 0.4. Daily Market Output. 6.24 kWh. CO2e per kWh. 立. 學. Increase rate i. 2%. ‧ 國. Decay rate. 政治 0.703 kg/kWh 大 2%. ‧. The Installation Cost includes the cost of a Tesla Powerwall ($6200) plus 24 solar panels with a. y. Nat. io. sit. unit price of $162 ($3888) and cables and racking ($1360). Finally, the set-up cost is estimated. n. al. er. to be $1100 including the personnel cost ($800) and the shipping cost ($300). The Maintenance. Ch. i n U. v. Costt of 200$ is a yearly estimate. At $12548, the installation cost is $ 2.01 per watt with 31%. engchi. being the cost of solar panels. The other variables in Table 1, like the Installation cost, are practical estimates. The characteristics of solar panels can be seen in the Appendix. They will be taken as basis for the analysis in Figures 6 to 8 unless they become independent variables.. For the price of Carbon Tax four scenarios are considered for Figures 6 to 8. The first one is PC02e = 0, no carbon price applied by the government. The second one is PC02e = 0.04, small carbon price applied by the government (40$/ ton of CO2e). The third one is PC02e = 0.08, an intermediate carbon price applied by the government (80$/ ton of CO2e). The fourth one is. 25.

(32) PC02e = 0.12, a high carbon price applied by the government (120$/ ton of CO2e).. The Figures 6 shows the impact of Carbon Tax on NPV under different levels of Increase rate (i). The Figure 6 will have fixed values established in Table 2 for Installation Cost, Length of Project (T), Maintenance Costt, Initial Price of Energy (P0), Daily Market Output, and Decay Rate. For the left panel, there is a high cost of investment (r = 0.08) and for the right panel, a low cost of investment (r = 0.02). The two variables are the Increase rate (i) and the Price of. 治政 Carbon Tax. For each panel, three scenarios for the Increase 大 rate and four for the Price of 立 Carbon Tax are crossed. With i = 0 there is no increase of the electricity price over the years. ‧ 國. 學. With i = 0.02 there is a small constant increase of the electricity price per year. With i = 0.05. Nat. y. ‧. there is a large constant increase of the electricity price per year.. er. io. sit. Figure 6: Impact of the Increase Rate and Carbon Tax (r = 0.08). n. al. i=0. i=0.02. $8,000. Ci=0.05 hengchi. iv n U i=0. i=0.02. i=0.05. $8,000. $6,000. $6,000. $4,000. $4,000. $2,000. (r = 0.02). $2,000. N $P V $-2,000. N $P V $-2,000. $-4,000. $-4,000. $-6,000. $-6,000. $-8,000. $-8,000 0. 0.04. 0.08. 0.12. 0. Carbon tax ($/kg CO2e). 0.04. 0.08. Carbon tax ($/kg CO2e). 26. 0.12.

(33) From the Figure 6, a positive and linear relation between the Increase Rate and NPV is observed. As the price of energy increases the savings induced by the solar panels increase as well. A positive and linear relation between the Price of Carbon Tax and variable NPV is also observed. A lower price of carbon induces a lower amount of Carbon Tax Avoided, therefore lower cashflows.. Another observation from the Figure 6 is that the project requires favorable investments. 政治大. conditions with a low discount rate r for the NPV to be positive and be economically viable. In. 立. the left panel, where r is high (r = 0.08), the NPV is negative for every factor tested. In the right. ‧ 國. 學. panel, where r is low (r = 0.02), the NPV is positive when the increase rate i is high (i = 0.05) whichever level of Carbon Tax implemented. When i is medium (i = 0.02) the break-even point. ‧. of the project is reach with a $ 59 tax per metric ton of CO2e emitted ($0.059/kg CO2e). n. al. er. io. sit. y. Nat. implemented.. i n U. v. The Figure 7 shows the impact of Carbon Tax on NPV under different Initial Prices of Energy. Ch. engchi. (P0). The Figure 7 will have a fixed value established in Table 2 for Installation Cost, Length of Project (T), Maintenance Costt, Increase Rate (i), Daily Market Output and Decay Rate.For the left panel, there is a high cost of investment (r = 0.08) and for the right panel, a low cost of investment (r = 0.02). The two variables are the Initial Energy Price (P0) and the Price of Carbon Tax. For each panel, three scenarios for the Initial Energy Price and four for the Price of Carbon Tax are crossed. With P0 = $0.20 the Initial Energy Price is considered low. With P0 = $0.40 the Initial Energy Price is considered intermediate. With P0 = $0.60 the Initial Energy Price is considered high.. 27.

(34) Figure 7: Impact of Initial Price of Energy and Carbon Tax (r = 0.08) P0=0.2. N P V. P0=0.4. (r = 0.02) P0=0.6. P0=0.2. $10,000. $10,000. $8,000. $8,000. $6,000. $6,000. $4,000. $4,000. $2,000. N P V. 立. $-4,000. $-10,000. $-2,000 $-4,000. $-8,000 $-10,000. 0.04. 0.08. 0.12. 0. Carbon tax ($/kg CO2e). ‧. 0. $-. $-6,000. ‧ 國. $-8,000. $2,000. 學. $-6,000. P0=0.6. 政治大. $$-2,000. P0=0.4. 0.04. 0.08. 0.12. Carbon tax ($/kg CO2e). er. io. sit. y. Nat. From the Figure 7, a positive and linear relation between the Initial Energy Price and NPV is. n. al. Ch. i n U. v. observed. A higher Initial Energy Price increases the savings on energy bills . A positive and. engchi. linear relation between the Price of Carbon Tax and NPV is also observed. A lower price of carbon induces a lower amount of Carbon Tax Avoided, therefore lower cashflows.. Another observation from the Figure 7 is that the project requires favorable investments conditions with a low r for the NPV to be positive and be economically viable. In the left panel, where r is high (r = 0.08), the break-even point is reached only when the P0 is high (P0 = $0.60) and with a $ 97 tax per metric ton of CO2e emitted ($0.097/kg CO2e) implemented. In the right panel, where r is low (r = 0.02), the break-even point is reach when P0 is intermediate (P0 =. 28.

(35) $0.40) and with a $ 59 tax per metric ton of CO2e emitted ($0.059/kg CO2e) implemented.. The Figure 8 shows the impact of Carbon Tax on NPV under different Length of Project (T). The Figure 8 will have a fixed value established in Table 2 for Installation Cost, Maintenance Cost, Initial Price of Energy (P0), Increase Rate (i), Daily Market Output and Decay Rate. For the left panel, there is a high cost of investment (r = 0.08) and for the right panel, a low cost of investment (r = 0.02). The two variables are the Length of project T and the Price of Carbon Tax.. 政治大. For each panel, three scenarios for the Length of Project and four for the Price of Carbon Tax. 立. are crossed under constant output performance. With T = 5 the project has a life span of five. ‧ 國. 學. years, after t = 5, the project does not generate cashflows anymore. With T = 10 the project has a life span of ten years, after t = 10, the project does not generate cashflows anymore. With T =. ‧. 20 the project has a life span of twenty years, after t = 20, the project does not generate. y. Nat. n. al. er. io. sit. cashflows anymore.. Ch. engchi. 29. i n U. v.

(36) Figure 8: Impact of Length of Project and Carbon Tax (r = 0.08) T=5. T=10. (r = 0.02) T=20. T=5. $2,000. T=20. $2,000. $-. $-. $-2,000. N P V. T=10. $-2,000. N P V. $-4,000 $-6,000. 政治大. $-6,000. 立 0. $-8,000 $-10,000. 0.04. 0.08. 0.12. 學. $-10,000. ‧ 國. $-8,000. $-4,000. 0. 0.04. 0.08. 0.12. Carbon tax ($/kg CO2e). ‧. Carbon tax ($/kg CO2e). sit. y. Nat. n. al. er. io. From the Figure 8, a positive and linear relation between the variable Length of project and. i n U. v. variable NPV is observed. A longer length of project increases the savings on energy bills. A. Ch. engchi. positive relation and linear between the variable Price of Carbon Tax and variable NPV is also observed. A lower price of carbon induces a lower amount of Carbon Tax Avoided, therefore lower cashflows.. Another observation from Figure 8 is that the project requires favorable investments conditions with a low r and Length of project T of 20 years for the NPV to be positive and be economically viable. In the left panel, where r is high (r = 0.08), the NPV is negative for whichever factor tested. In the right panel, where r is low (r = 0.02), the break-even point is reached when T = 20 and with a $ 59 tax per metric ton of CO2e emitted ($0.059/kg CO2e) implemented. 30.

(37) Moreover, the slopes of the trendlines in Figure 8 are different than the ones in Figure 6 and 7. While the slopes of the trendlines in Figure 6 and 7 are the same for each factor tested, in Figure 8 the slopes are increasing with T. It can be said that the Price of Carbon Tax and Length of Project T have a positive interaction effect on NPV under constant output performance degradation. That is, the economic benefits of solar panels installation are more salient when Carton Tax and project duration are both high.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 31. i n U. v.

(38) 6. Discussion & Conclusion. Research showed that acting toward mitigating the global warming issue was compulsory or mankind will face unbearable financial, social and environmental costs. Solutions exists to reduce the greenhouse effect. Among them the Green Logistics practices such as implementing solar panels on a warehouse’s roof. The three factors tested in our NPV model, the increase rate. 政治大 to the NPV of the project. The Carbon Tax level, tested in function of the three other factors was 立 of energy, the initial price of energy, and the time length of the project were all positively related. ‧ 國. 學. also positively related toward the NPV. With that said, when the discount rate is high, NPV values of Solar Panels Installation are nearly all negative in tested scenario. Keeping the. ‧. discount rate low would mean a low cost of investment. A special credit rate for loan to green. sit. y. Nat. projects could drive the discount rate down and favorize investments in solar panels installation. io. al. n. help companies to invest in Green Logistics project.. Ch. engchi. er. on warehouse’s roof for example. It would be beneficial for both companies and the planet to. i n U. v. Nevertheless, the model suffers from a few limitations. The data taken for the analysis were for small scale photovoltaic (PV) projects, like household and little business infrastructure. Data for bigger scale of projects would have been preferable but were not available or in a less precise way. Non-residential solar installations, because significantly bigger, benefit from an economy of scale compared to PV residential installations. Therefore, the economy of scale could not be included in our model while an installation on a warehouse or plant’s roof is considered a non-residential solar installation. Regarding the Daily Market Output, a link with the Installation Cost could have been establish to consider different sizes of project of solar. 32.

(39) panels implementation, assessing an optimal size for an economic performance. Moreover, the leasing of the warehouse’s roof to a solar company, another mode of investment for solar panel installation was omitted in the study. This different business model put the investment weight on the photovoltaic supplier instead of the company owning the warehouse.. The model presented in this paper has put forward the economic and environmental performance through the NPV model, which gives the value in dollar and the reduction of. 政治大. CO2e in kilograms. However, the social performance is not be considered in the model. It is a. 立. significant factor but the quantitative analysis performed later is limited and cannot take it. ‧ 國. 學. into account. Moreover, the NPV model can be apply to other business fields than Logistics. For example, a production company can implement solar panels on a roof of a plant. The. Nat. sit. y. ‧. model can also be adapted to other type of projects.. io. er. Despite those limitations, the study delivers practical contributions. Based on the analysis in. al. Section 5, a high price of energy favorizes the NPV of the implementation of solar panels.. n. v i n Places where the energy is hardC to h get and expensive U e n g c h i would better benefit from this type of. project, as well as locations where the price of energy is expected to increase. As seen in section 5, a certain Carbon Tax level help to reach the break-even point when conditions do not suffice on their own. Under the estimated conditions in Table 2 and a discount rate of 2%, $ 59 dollars per metric ton of CO2e emitted would turn the NPV positive and make the project economically viable. Currently few countries implemented such Price of Carbon Tax. The trend is rising though, with the increasing need of a Carbon Tax to mitigate the consequences of the global warming issue. A higher Carbon Tax benefits more with a longer length of project. In the case of a solar panel installation, product characteristics should fit a long lifespan of project to. 33.

(40) get both best economic and environmental value.. Future studies could expend the work by incorporating randomness into model elements, e.g. uncertainty in the Price of Carbon Tax, Decay Rate or Increase rate of energy price. The size factor of the solar panel implementation could be evaluated to estimate the best suitable size for economic viability of the project. Other types of Green logistics projects could be covered using an adapted version of the model to assess their financial feasibility.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 34. i n U. v.

(41) Reference. Aldy, J. (2017). Designating and Updating a US Carbon Tax in an Uncertain World. Discussion Paper 17-01: Resources for the Future, pp.4-21.. Barbose, G. and Gargouth, N. (2016). Tracking the Sun IX: The Installed Price of Residential. 政治大. and Non-Residential Photovoltaic Systems in the United States. Energy Technologies Area,. 立. Lawrence Berkeley National Laboratory, p.16.. ‧. ‧ 國. 學. Brander M., (2012) Greenhouse Gases, CO2, CO2e, and Carbon: What Do All These Terms. y. Nat. n. al. er. io. sit. Mean? Ecometrica, pp.1-3.. i n U. v. Carter, C. and Rogers, D. (2008). A framework of sustainable supply chain management:. Ch. engchi. moving toward new theory. International Journal of Physical Distribution & Logistics Management, 38(5), pp.360-387.. Cook, B., Smerdon, J., Seager, R. and Coats, S. (2014). Global warming and 21st century drying. Climate Dynamics, 43(9-10), pp.2607-2627.. Doraszelski, U. (2001). The net present value method versus the option value of waiting: A note on Farzin, Huisman and Kort (1998). Journal of Economic Dynamics and Control, 25(8),. 35.

(42) pp.1109-1115.. Emmett, S. and Sood, V. (2010). Green supply chains. Chichester, U.K.: John Wiley & Sons.. Hansen, J., Sato, M., Hearty, P. and Others (2016). Ice melt, sea level rise and superstorms: evidence from paleoclimate data, climate modeling, and modern observations that 2 A degrees C global warming could be dangerous. 1st ed. eScholarship, University of California, p.3762.. 立. 政治大. IPCC (2007). IPCC Fourth Assessment Report: Climate Change 2007. Intergovernmental. ‧ 國. 學. Panel on Climate Change. Cambridge University Press, Cambridge.. ‧. IPCC (2015). Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II. y. Nat. sit. and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. al. n. p.13.. er. io. [Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, 151. Ch. engchi. i n U. v. Landrigan, P. (2017). Air pollution and health. The Lancet Public Health, 2(1), pp.e4-e5.. Le Monde.fr. (2017). Climat : Paris vise la neutralité carbone en 2050. [online] Available at: http://abonnes.lemonde.fr/planete/article/2017/03/13/climat-paris-vise-la-neutralite-carbone-e n-2050_5093437_3244.html?h=8 [Accessed 24 Jun. 2017].. Lin, Grier C. I.; Nagalingam, Sev V. (2000). CIM justification and optimisation. London: Taylor & Francis. pp. 36. ISBN 0-7484-0858-4. 36.

(43) McDonald, R. and Siegel, D. (1986). The Value of Waiting to Invest. The Quarterly Journal of Economics, 101(4), p.707-728.. Rezai, A., Van der Ploeg, F. and Withagen, C. (2012). The Optimal Carbon Tax and Economic Growth: Additive versus Multiplicative Damages. Center for Energy and Environmental Economic Studies. St. Petersburg: EUSP, pp. 1-27.. 政治大. Rotemberg, J. and Woodford, M. (1996). Imperfect Competition and the Effects of Energy. 立. Price Increases on Economic Activity. Journal of Money, Credit and Banking, 28(4),. The Economist. (2017). Triple bottom line. [online] Available at:. Nat. y. ‧. ‧ 國. 學. p.549-577.. n. al. er. io. sit. http://www.economist.com/node/14301663 [Accessed 24 Jun. 2017].. i n U. v. US EPA. (2017). Greenhouse Gases Equivalencies Calculator - Calculations and References | US EPA. [online] Available at:. Ch. engchi. https://www.epa.gov/energy/greenhouse-gases-equivalencies-calculator-calculations-and-refer ences [Accessed 3 Jun. 2017]. van den Bergh, J. (2004). Optimal climate policy is a utopia: from quantitative to qualitative cost-benefit analysis. Ecological Economics, 48(4), pp.385-393.. Zhang, K., Douglas, B. and Leatherman, S. (2004). Global Warming and Coastal Erosion. Climatic Change, 64(1/2), pp.41-58. 37.

(44) Appendix. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 38. i n U. v.

(45)