Hacia el Futuro: Energy, Economics, and the Environment in 21 st Century Mexico

1. Hacia el Futuro: Energy, Economics, and the Environment in 21st Century Mexico Maria E. Ibarraran and Roy Boyd EPA and SMF Workshop Rio de Janeiro, September 2006

2. Content of the book Part I 1. Introduction 2. Greenhouse Gas Emissions and Climate Change 3. Forecasting the Impact of Climate Change 4. Energy Use in Mexico (and Latin America) 5. Economic Theory, Emissions Control, and Kyoto

3. Content (cont) … and emphasis of what our model does Part II 6. A Dynamic General Equilibrium Model 7. Simulation Results Under Competitive Scenarios 8. Simulation Results Under Imperfect Market Scenarios 9. Simulation Results with Emissions Trading 10. Conclusions

4. Special features of our work • Combines science, economics, and policy-making • Science of climate change and evidence • Emissions worldwide, in Latin America and in Mexico • Regional and local impacts, especially for Mexico • Energy use and trends in Mexico, Brazil, Argentina and Venezuela • Economic analysis of climate change and possible solutions through incentives • International agreements

5. Table1. Carbon emissions from fossil fuel consumption and flaring, (million metric tons of CO2) Source: EIA, 2002.

6. Expected impacts of climate change for Mexico • Regional Impact • Specific vulnerability • Agriculture • Forest ecosystems • Desertification and Drought • Hydrology • Coastal zones • Human Health

7. Expected effects by 2025-2050 Corn production scenario Desertificationscenario Sea level rise scenario Drought scenario

8. Energy Use Primary energy supply by fuel type Total final energy consumption by sector

9. Energy Prices Relation between domestic prices and opportunity costs or external prices for petroleum fuels and natural gas (1970-1988) Comparison of electricity rates between Mexico and U.S. (domestic rate/US rate)

10. Source: EIA, 2004.

11. Features of the Dynamic Computable General Equilibrium Model • The model is calibrated using 2000 data and is run for a total of 21 years to 2020. • The model is composed of 9 production sectors 7 consumption sectors, 4 consuming agents segregated by income group, a government sector, and a foreign trade sector. • The model solves in such a way that savings plus imports are equated with investment plus exports.

12. Features of the Dynamic Computable General Equilibrium Model (cont.) • The model assumes rational expectations on the part of all agents • Labor and productivity growth are consistent with present OECD projections (i.e. 1.3% and 1.6%)

13. Further Model Assumptions • All demand and substitution elasticities used are based on current empirical estimates and varied in a sensitivity analysis. • A flexible nesting structure is used based on a CES production and consumption structure. • The model has special features to all for taxes, tariffs, subsidies, technological change and depletion. • The model is constructed in such a way as to allow detailed analysis of the fossil fuel and power sectors. • A total of 25 simulations were run assuming a variety of policies and market conditions

14. Consumer Goods and Producing Sectors Production Goods Services 1. Manufacturing Manufacturing Goods 1. Food 2. Coal Mining Coal 2. Energy 3. Chemicals and Chemicals and Plastics 3 . Autos Plastics 4. Agriculture Agricultural goods 4. Gasoline 5. Services Producer Services 5. Consumer Transport Transportation for 6. Transportation 6. Consumer Services production 7. Housing and 7. Electricity Electricity Household goods 8. Oil and Gas 1. Crude Petroleum 2. Natural Gas 9. Refining output Refined output Sectors in the Model

15. Mathematical Description of the Model s s ( / - 1) + f s s s s s s ( - 1)/ ( - 1)/ ( - 1)/ = [ + d d d ] 1) V L K M t t t t L K M t d d ? , , > 0 and L K M d d ? + + = 1 L K M For each household c total utility is modeled by t he function, - t S r (2) U = U (X , R ) * (1+ ) t = 1, …, n c t c,t c,t c,t

16. Each consumer’s expenditure co nstraint can be written as, n å (3) (TG + TF + (P * L ) + (r * K * S )) = c,t c,t L,t c,t t c,t = t 1 n å ((INV * S ) + (P * X ) + (P * R )) t c,t I,t c,t L,t c,t = t 1 where endowments are given on the left - hand side of the equation and expenditures are placed on the right hand side. TG and TF represent the c,t c,t transfer to the consumer from the government and from the foreign agents, P is the tax exclusive price of labor and r is the rental rate of capital. K is L,t t the level of capita l stock in period t , S is the share of total capital owned by c,t consumer c, INV is the total investment in time period t , and P is the tax t I,t inclusive vector of prices for consumer goods = a a a a G Ax x x x 1 2 i n u 1 2 i n å a = 1 (4) i i 1 n Õ = a E P i i A = i 1

17. The balance of trade relationship is given by, S S S ( 5 ) (P * IM ) = (P * EX ) + TF t = 1, ..., n m,t j,t j,t j,t c,t where IM is a (nine dimensional) vector representing the quantity o f each of j,t the producer goods imported, P is the vector of imported goods prices, EX m,t j,t is the vector of producer goods exported, P is the tariff inclusive vector of j,t producer goods prices, and TF is the level of foreign transfers which can be c,t posit ive, zero, or negative. More formally, the growth in the effective labor force over time is given by the equation, g (6) L = L (1+ ) t+1 t g where is the composite of the growth rate of population over time and the growth in the effectiveness of the typica l worker

18. The capital growth rate is modeled in accordance with capital theory and is represented by a system of three equations. For each time period t we have, ( 7 ) P = P t = 1, ..., T A,t k,t+1 where P is the weighted (aggregate) tax exclusive price of consumption, A,t (i.e. the weighted average of the P ’ s) and P is next year’s tax exclusive I,t k,t+1 price of capital. This says that the opportunity cost of acquiring a unit of capital next year is a unit of consumption in the present period. We also have ( 8 ) P = (1+r ) P t = 1, ..., T k,t t k,t+1 meaning that the price of capital in this period, P , must be equal to the k,t coming period’s rental value of capital, r *P , plus next period price of t k,t+1 capital, P . Finally, we have k,t+1

19. Following Lau, Puhlke, and Ruthe rford (1997) we divide the problem into two distinct sub - problems, one defined over the finite period from t = 0 ¥ to t = T and the second the infinite period from t = T+1 to T = . Hence, the first problem is 1 T S t Max ( ( , ) ) ( 10 ) U X R c, t c, t c, t r 1 + t =0 subject to T T å å - = L + (1 1 ) S S P X P P K P K c , t A , t c , t L , t k , 0 c,0 k, T + 1 C , t C , T + 1 c, T + 1 t =0 t =0 (9) Kt+1 = Kt(1-Δ) + INVt t = 1, …, T where Δ stands for the rate of depreciation and INV stands for gross investment. This states that the capital stock in the next period must be equal to this year’s capital stock plus net investment. Taken together, equations 7-9 insure that economic growth will be consistent with profit maximizing behavior on the part of investors.

20. and = + (1 1 a) for all t = 0, 1, ... T L R L c, t c, t c, t and the second problem is ¥ 1 å t Max ( ( , ) ) (1 2 ) U X R c, t c, t ct r 1 + t = T + 1 subject to ¥ ¥ å å . = + (1 3 ) S P X P L P K I, , t c, t L, t c, t K, T + 1 c, t + 1 c, T + 1 t = T + 1 t = T + 1 = + (1 3 a) for all t=T+1, ...? L R L c, t c, t c, t where ? is the rate of time preferences, r and K refer to the rental value of o c,o K capital and quantity of capital before the terminal period, r and refer T+1 c,T+1 L to these variables after the terminal period, and is total labor plus leisure c,t th for each agent in the t time period. P stands for the tax exclusive price of K, t capital, and , as before, P and P stand for the tax inclusive price of I,t L,t consumer goods and the tax exclusive price of labor respectively.

21. we include the level of post - terminal capital as a variable and add a constraint on investment growth in the final period. Thus we have (1 4 ) INV /INV = Y /Y T T - 1 T T - 1 where Y gives GDP at time T . This constraint imposes a balanced growth in T the final period, but does not require that the model achieve steady - state growth.

22. Perfect Competition • The model is first run under the assumptions of perfect competition and full employment to see the impact of various energy policies and technological change on key economic variables as well as CO2 emissions. • A total of 9 different scenarios are run under these assumptions.

23. Scenario 0 Scenario 0 : The benchmark case : The benchmark case Scenario 1 Scenario 1 : The benchmark case plus oil depletion : The benchmark case plus oil depletion Scenario 2 Scenario 2 : Scenario 1 plus deregulation of energy prices : Scenario 1 plus deregulation of energy prices Scenario 3 Scenario 3 : Scenario 1 plus new investment in PEMEX and CFE producing : Scenario 1 plus new investment in PEMEX and CFE producing capital capital - - enhancing technological change in energy sectors enhancing technological change in energy sectors Scenario 4 Scenario 4 : Scenario 1 plus deregulation of energy prices and capital : Scenario 1 plus deregulation of energy prices and capital - - enhancing technological change in energy sectors enhancing technological change in energy sectors Scenario 5 Scenario 5 : Scenario 2 plus energy efficient technological change in all s : Scenario 2 plus energy efficient technological change in all s ectors ectors Scenario Scenario 6 6 : : Scenario Scenario 4 plus a 4 plus a carbon carbon tax tax Scenario Scenario 7 7 : : Scenario Scenario 5 plus a 5 plus a carbon carbon tax tax Scenario 8 Scenario 8 : Scenario 3 plus a carbon tax and capital : Scenario 3 plus a carbon tax and capital - - enhancing technological enhancing technological change in all sectors change in all sectors

24. 14 12 10 8 6 4 2 0 Sc 0 Sc 1 Sc 2 Sc 3 Sc 4 Sc 5 Sc 6 Sc 7 Sc 8 GDP CO2 Emissions GDP and CO2 Emissions Under the Various Scenarios Run

25. Comparison of Scenarios Scenario 1 is compared to Scenario 0 Scenario 2 is compared to Scenario 1 Scenario 3 is compared to Scenario 1 Scenario 4 is compared to Scenario 1 Scenario 5 is compared to Scen ario 4 Scenario 6 is compared to Scenario 4 Scenario 7 is compared to Scenario 6 Scenario 8 is compared to Scenario 6

26. Scenario1 Scenario2 Scenario3 Scenario4 Scenario5 Scenario6 Scenario7 Scenario8 GDP -1.81% 0.06% 1.13% 0.92% 1.25% -0.19% 0.97% 18.43% Final level of Investment -11.18% 0.76% 3.75% 2.99% 4.01% -4.32% 0.00% 19.03% Oil output -37.09% -0.21% 31.64% 31.43% -28.30% -14.62% -25.38% 5.87% Power output -2.72% -0.48% 15.64% 15.06% -3.02% -3.78% -12.64% 16.39% Consumption 1.42% -0.05% 0.29% 0.23% 0.39% 0.08% 0.17% 15.46% Imports -0.01% 0.00% 0.01% 0.01% -0.01% -0.01% -0.02% -0.06% Exports -3.65% -0.13% 2.34% 2.21% -2.16% -1.38% -1.73% 6.50% Exports oil -37.13% -0.15% 31.92% 31.77% -28.14% -14.56% -24.85% 1.55% BoP surplus -45.41% -2.82% 51.12% 48.30% -32.65% -20.84% -32.33% 123.72% Cumulated welfare agent 1 -0.34% -0.10% 0.43% 0.33% 0.02% -0.07% -0.14% 12.46% Cumulated welfare agent 2 -0.30% -0.09% 0.43% 0.34% 0.02% -0.07% -0.14% 12.59% Cumulated welfare agent 3 -0.75% -0.06% 0.21% 0.15% 0.03% 0.02% -0.04% 8.69% Cumulated welfare agent 4 -1.02% 0.08% -0.04% 0.04% 0.06% 0.14% 0.10% 4.56% Terminal capital stock -4.37% -0.36% 1.07% 0.71% 1.94% -1.68% 0.46% -0.72% Cumulated Govt. revenue from PEMEX -10.20% 0.32% -4.55% -4.22% 1.02% -4.41% 3.19% -3.19% Cumulated Govt. revenue from CFE 2.33% 0.00% -2.27% -2.27% 6.98% 0.00% 2.33% 20.93% Cum. Govt revenue from other sources 1.29% -0.02% 1.40% 1.38% 0.50% 2.55% -0.30% 44.84% CO2 Emissions -31.98% -0.25% 27.90% 27.75% -27.76% -14.18% -24.69% 9.28%

27. Important Points From these Simulations • Our results make it clear that discussions of energy policy have to recognize the importance of depletion and investment in the choice of energy alternatives. • For every ton of carbon emissions avoided, GDP declines by about $104 dollars. (note-very much in line with other’s results e.g. EPA) • This number is fairly high with respect to other developing countries (i.e. China) and reflects the fact that Mexico has little ability to shift from high carbon content fuels (such a significant coal deposits) to fuels with substantially lower or no carbon content (such as natural gas, hydroelectric power, or nuclear power.

28. Important Points (Cont.) • The results of our simulations show that a carbon tax, in spite of the significant environment benefits in terms of stemming carbon dioxide and other harmful emissions, will entail significant costs in terms of both economic efficiency and consumer equity • The simulations so demonstrate the importance of technology for both the energy and non energy sectors. • Our results are quite robust with respect to the parametric assumptions made in terms of the speed of growth and the elasticities of demand as well as substitution in production.

29. Results Under Imperfect Competition and Non-frictional Unemployment • The model is then run assuming that we have sticky wages to see the effects of this on our modeling results. • We then run the model under the assumption of monopoly power in the petroleum (and petroleum products) industry. Similar assumptions are used for the power industry.

30. Scenario 9 : Sticky wages in labor markets, deregulation of energy prices, andcapital - enhancing technological change in energy sectors Scenario10 : Scenario 9 without capital - enhancing technological change in energysectors Scenario11 : Scenario10plusenergyefficient technologicalchangeinall sectors Scenario12 : Scenario 9 plus capital - enhancing technological change in all sectors and a carbon tax Scenario13 : Monopoly power in energy sector, deregulation of energy prices, and capital - enhancing technological chan ge in the energy sectors Scenario14 : Scenario 13 with energy efficient technological change in all sectors, instead of capital - enhancing technological change in the energy sectors Scenario15 : Scenario 13 without any technological change Scenario16 : Scenario 12 plus monopoly power in the energy sectors

31. 16 14 12 10 8 6 4 2 0 Sc 0 Sc 9 Sc 10 Sc 11 Sc 12 Sc 13 Sc 14 Sc 15 Sc 16 GDP CO Emissions 2 GDP and Emissions for Scenarios Nine through Sixteen

34. Important Points From the Second Set of Simulations • Technological change is essential for sustained growth. • The nature of the technological change, however, is critical in terms of the environment. • Carbon tax design is a great importance. • The effect of Monopoly is ambiguous. • Persistent unemployment combined with carbon taxation can have disastrous effects.

35. Emission Trading with the U.S • The model is re-aggregated to allow us to see the workings of the forestry and agricultural sectors. • It is then re-run under the assumption that emitters of GHG’s have to purchase carbon sequestration “rights” in order to emit over a certain specified level. • Initially the analysis is limited to emitters and forest owners within Mexico.

36. Emission trading with the U.S. (continued) • The model of Mexico is then combined with a similar dynamic CGE model of the U.S. • The model simulations are then re-run assuming that emitters from the U.S. are required to purchase sequestration “rights” from forest owners in Mexico. • The results of the two sets of simulations are then compared to see what is most cost effective.

38. Conclusions with Emissions Trading • If (after some level of emissions) GHG emitters are required to purchase sequestration rights for a certain percentage of their additional carbon emissions, such rights can lower GHG levels both by sequestering significant levels of carbon in carbon sinks and by reducing the level of the producers’ GHG emissions. • The costs are relatively high if trading is combined to Mexico. • When trades between the U.S. and Mexico are allowed, however the situation changes and the aggregate costs decline by over 80%.

39. Emissions trading conclusions (continued) • When permits are required, U.S. producers cut the usage of all fossil fuels but do so relatively more with coal and oil than on natural gas. This is impossible for Mexican emitters to do and underscores the importance of international permit trading. • Such gains come at relatively low costs when countries with ample reserves of low carbon fuels, such as the U.S., can participate. • The potential for unintended spillovers such as carbon leakage seem to be relatively small in scope and easily managed with proper policy design.

40. General Conclusions • Two features (i.e. depletion and technological change) to a large extent drive the overall results of our simulation analysis in terms of emissions. The relationship of the two is complex but critical to our analysis. • An overall finding of our analysis is that energy efficient technological change is of major importance if Mexico is to seriously reduce emissions without experiencing harmful effects on its economic growth and welfare.

41. General Conclusions (continued) • It is also essential to eliminate market distortions such as energy price distortions and labor market imperfections if policymakers want to guard against severe economic contraction when carbon taxes or other similar sorts of emission controls are introduced. • Finally, although the exercise of monopoly power by state owned energy producers has the potential to decrease emissions through supply restrictions, the use of such power is not to be recommended.

42. …and some recent work • Simulate the effect of extreme weather events on the Mexican economy: drought • We use model used for emissions trading • Scenarios cover: • No drought • Severe drought • Adaptation policies (agric., forestry, power)

44. …and basic results • Drought has a sectoral impact, mainly on agriculture and grazing, forestry, and power generation (hydro), and ripple effects throughout the economy • Regressive impact on welfare • Adaptation policies (technological change and irrigation) can only partly mitigate its effects.