Updated: June 6, 1999

[dont forget to refresh your browser]

DIRECTIONS: The following assignments should be typed with a word count indicated at the top of the first page. Use the word processor to find the word count [eg, in WORD, "Word Count" under "Tools"]. Exclude tables, footnotes and references from your count. The nominal length is around 1000 words, but you can go over if it is really worthwhile. You can also go under if you can make your point using less verbiage. Think of it this way, the marginal cost in terms of a grade penalty is zero up to 1000 words; after that the marginal cost starts to increase, and at an increasing rate. If the marginal benefit of extra words is higher than the marginal cost, then it pays you to expand the length; if not, shrink the length. I am a bit ambiguous on what the marginal cost is, exactly; but I will always ask if you could have communicated the same message with less space.

Please do not hand in assignments electronically. I would like to receive a typed paper copy.  

For the most part, your assignments are in the form of memos to your boss. Your boss is a busy woman who heads the US negotiating team on climate policy. This includes being in charge of further negotiations on the Kyoto protocol. Think of her as a former Stanford grad who majored in economics with a natural science minor but who graduated 10 years ago and has forgotten many of the details.

1. Due April 5: Your boss has heard that there is some uncertainty about how real the climate change problem is. Write a short paper indicating the extent to which climate change is a real problem and why there is uncertainty about the natural science dimensions of the problem. Feel free to do your own analysis, though original analysis is not necessary. If you do analysis, you may wish to estimate the statistical relationship between global average temperature and the stock of carbon dioxide.

2. Due April 7: Calculate how much CO2 you personally were responsible for in 1998. Include emissions that you directly caused (by, for instance, burning natural gas), as well as emissions that you induced, by purchasing goods produced with carbon dioxide emissions.

3. Due April 12. Ajax Mining and Edison Power both generate pollution, A and E. Their production costs actually decline the more they pollute. Supressing the use of inputs and goods output, costs can be expressed as the downward sloping portion of C_A(A) = 2(A-10)²-5 and C_E(E) = 3(E-15)²-4. Damage from pollution is quadratic: D(E+A)=(E+A)².

(a) Uncontrolled, how much will each firm emit?

(b) Write a GAMS program to calculate the emissions from each firm that minimize total social costs (costs to the firm plus pollution damage). Attach a printout of the GAMS program, as well as the GAMS results.

(c) Solve the problem using calculus and compare your answer with part (b).

 4. Due April 19. You still work for the head of the US climate negotiating team. Write a memo to your boss indicating what the economically efficient worldwide level of carbon control is. Use your own version of the DICE model to generate your answers but report to your boss in a non-technical manner with supporting tables if necessary (in other words, explain what your results are). Indicate what caveats should be placed on the results of your analysis. The marginal cost per word is zero up to 1000 words (excluding tables, footnotes and references). [Hint: you can type in DICE from the back of the Nordhaus book or download it from his webpage, remembering to delete commas in comments and to keep * out of column 1 unless the line is a comment (http://www.econ.yale.edu/~nordhaus/homepage/manageapp.htm).]

5. Due April 28. Your boss asks you to clarify for her the significance of discounting to the US policy position on climate change. She has heard some argue for high discount rates, others for low discount rates. She has heard some speak of hyperbolic discounting and others of exponential discounting. Write her a memo to help clarify this issue for her, indicating how significant this issue is, particularly in terms of optimal levels of emission control in the present. Be sure to cite the literature as well as your own analysis of the problem. The marginal cost per word is zero up to 1000 words (excluding tables, footnotes and references).

6. Due May 5. Your boss asks you how important technological change is to greenhouse policy. Some constituents have been advocating strict carbon regulations in order to induce dramatic reductions in costs of controlling carbon, though the cost reductions will not be seen for several decades. Thus, eventually, carbon emissions will be dramatically reduced, primarily by making it very cheap to do so. Others have suggested that additional technological change brought on by strict carbon regulations is unlikely to make much difference in climate control costs over the next few decades. Your boss in confused and wants you to clear the air for her and advise her as to whether she should push for strict climate regulations in order to induce technological change. [If you plan to use DICE for this assignment, you may wish to check out some notes on the subject.]

7. Due May 12. Your boss has heard that there is a great deal of uncertainty about the damage from climate change. She is worried that your recommendations that only modest controls be adopted (in your memo of April 19) are driven by your assumptions about damage (which she reminds you was 1.33% of world GDP loss from a 3^oC temperature rise). She has read that in Bill Nordhaus' survey of experts (discussed in his book), experts differed on their assessment of the consequences of a 3^oC temperature rise. The median response was 1.9% of world GDP and the mean response was 2.9%. Some responses were clearly much higher (and much lower) than these. For larger temperature changes, the mean damage was 6.1%. All this is rather unsettling to your boss. She has been basing the administration's position on your recommendation that emissions should be about 9% lower than they otherwise would be. She is proposing supporting this with a global carbon tax. She is getting heat for this. She knows many people are saying the consequences of global warming may be very severe. She would like to know how much the policies she is suggesting for the very near term would be changed if you are wrong about the magnitude of damage. She wants to know the effect of different assumptions about damage on near-term emission control levels, as well as the effect on the carbon tax that she is pushing to implement the emission control policy. [Hint: Use DICE among other sources; the carbon tax is the marginal rate of substitution between carbon and output, which can be computed using the shadow prices on the carbon constraint and the consumption constraint.]

8. Due May 20. Your boss is still worried about the uncertainty in damage from climate change. She is considering financing a crash research program to resolve some of this uncertainty. She wants to know what kind of research funding might be justified. For instance, she asks you if we knew now what the damage was from climate change, how much better off would we be relative to the case of staying in the dark, with damage remaining uncertain forever? She asks you to help decide on the appropriateness of the R&D funding, providing her guidance. She knows you have a climate model that can give you an answer to this although she doesn't trust models. She is interested in a quantitative answer from your analysis but she also wants an explanation and justification of your recommendations without relying on the model. You may assume that the current level of uncertainty regarding damage is that there are two possible states of the world for damage, high and low. Low damage means that a 3^oC temperature change would result in no loss of GDP; high damage means a 6% loss in GDP. The probability of high damage is 0.22 and the probability of low damage is 0.78, for an expected loss of 1.3% of GDP. [Hint: Strictly speaking, the uncertainty case should involve writing DICE's objective function as expected utility, with variables depending on the state of the world. One perfectly reasonable approximation is to figure out the control levels when damage is at the mean of 1.3%. Then see what the net benefits are for 6% damage and the benefits are for 0% damage, holding control levels at those for the mean of 1.3%. Then take the expectation of these benefits. A slightly simpler approach is to assume that these are small effects and thus that control and benefits under perpetual uncertainty are approximately the same as control and benefits with damage at the average of 1.3%. Thus the perpetual uncertainty case is the certainty case you have already run in previous weeks. Remember that the objective function is in utility units; you will need to examine differences in the net present value of consumption (or output), using the endogenously determined consumption discount rate or by converting utility to consumption using the marginal utility of consumption.]

9. Due May 26. Your boss has read Bill Nordhaus' paper proposing a climate allowance protocol. She is concerned about two items in the proposal. One is enforcement. She is not convinced that the countervailing tariffs will work. She thinks they need to be spelled out more specifically; furthermore, she must be convinced that the cost of the countervailing tariffs will exceed the gains from defecting from the protocol and that the tariffs are not too onerous on the imposing country. A second problem she has is with the mechanism for determining the overall control rate by querying willingness to pay of participating countries. She doesn't understand this procedure completely and she doesn't understand why Nordhaus claims it is in the best interest of countries to truthfully reveal their willingness-to-pay. She knows you are a busy camper and probably can't answer both of these questions. But she would like you to address one of them.

FINAL EXAM

Directions: Write a paper answering the question; bring it with you to your scheduled time on Monday, June 7. The paper should be very carefully written and concise. At your exam time, be prepared to give a polished five minute presentation of your paper. As always, type the number of words at the top of your paper. Remember, if you go over 1000 words, make it worthwhile.

FINAL EXAM QUESTION:

Somewhat surprisingly, Congress has ratified the Kyoto Protocol. Your boss has picked up the Administration's analysis of the protocol (the Council of Economic Advisors' analysis from July 1998) and notes that we had indicated we would adopt some sort of "mandatory domestic emissions trading system" to meet the Kyoto targets. Your boss needs a proposal on her desk Monday morning of just such a system. She wants you to design the system, indicating, among other things, how permits would be initially distributed/sold, how they would be traded, across space and time, how compliance with the program would be verified and enforced, and -- all the details. She also wants your assessment of the political feasibility, economic efficiency and distributional consequences of your proposal. Be quantitative and concise. Try to anticipate some of the problems that are not obvious from a cursory examination of such a proposal. Demonstrate your skills as an economic analyst and don't spend too much time reviewing the literature. She wants answers.

You will be graded on how complete your proposal is (ie, have you thought of all the details that are important in the real world), how innovative/creative your proposal is, how much you demonstrate your mastery of the economics of the problem, how aware you are of incentive effects, how well you have designed your proposal to reflect the political economy dimensions of regulation (ie, the political forces unleashed by the monetary burdens and benefits of your proposal). Most of all, you should demonstrate your skills as an economic analyst. In your presentation, be professional and to the point. Finally, this should be your own work. Do not consult others. You may, however, use any written materials you may find useful. The usual length limit applies (1000 words unless there is a good reason for more).

GOOD LUCK!