American Enterprise Institute

June 28, 2007

[Edited transcript from audio tapes]

1:45 p.m.	Registration
2:00	Presenter:	W. David Montgomery, CRA International
	Discussants:	Dallas Burtraw, Resources for the Future
		Lee Lane, CRA International
	Moderator:	Samuel Thernstrom, AEI
4:00	Adjournment

Proceedings:

Samuel Thernstrom:  Good afternoon and thank you all for coming.  If everyone can get settled, I think we can get under way.  This should be really an exceptional event.  In fact, I think of all the climate change events I have been involved with at AEI, I personally am more interested in hearing today’s speakers than practically any other event I have been involved in.  So I’ll try to speak only very briefly and move along to the main event without further ado.

As all of you note, this is a particularly decisive time for American climate policy.  Congress is wrestling with various proposals to regulate greenhouse gas emissions in one way or another.  I think most of us on this panel believe that it is quite likely that some form of a national cap-and-trade program, or at least an emissions limitation program, is quite likely to be enacted sometime in the next few years, anyways.  Meanwhile, there is also a great deal of action on the state, local and regional level as well. 

There are reportedly 300 different climate bills under consideration on the state and national level right now.  These proposals are being hotly debated, in part because of the precedent-setting nature of climate policy at this time.  Everyone involved in these debates expects that the basic direction of American climate policy will be determined in the next few years.  And state level initiatives, such as California’s, are widely expected to, and indeed are, deliberately constructed to influence the shape of the future federal policies.

So these state climate policies are obviously very important, and there is a lot that we can learn from them.  One of the great advantages of America’s federal system obviously is that states are able to experiment with their own approaches to complex public policy questions and Congress can look to the successes and failures of these state laws when considering its own options.

So our question today is what we can learn from California’s efforts to reduce greenhouse gas emissions.  California, of course has, I think, deserved reputation as the nation’s greenest state.  But many of us also remember that it had some rather serious problems with its electricity pricing and supply just a few years ago.  And so, one question I think we will consider today is what the interplay between those different considerations may be in the coming decades.

 As most of you here may know, California Governor Arnold Schwarzenegger signed California’s Climate Bill into law September 27th, 2006.  The law requires California to cap its greenhouse gas emissions at 1990 levels by the year 2020.  This is the first law in the country establishing a mandatory cap on statewide greenhouse gas emissions, and it is, I think it is fair to say, an exceptionally ambitious target, particularly given that the state’s population is expected to grow by 42 percent during that same time period.

Now the California law requires very aggressive emissions cuts but it leaves key implementation decisions to the State Air Resources Board and other agencies.  The Governor and the legislature have already clashed over these questions, and I think we will hear something today that will clarify what the particular options may be that they will have to decide on in the next few years.

And the stakes here are obviously very high as I have just described, and there are important and complex questions to consider.  Can California even meet these ambitious environmental goals; and if so, at what cost?  If California succeeds in this effort, that will be a remarkable achievement.  And if it fails, many critics of greenhouse gas regulations can usually point to it as justification for taking a very different approach with national policy.

We are very fortunate to have three exceptional scholars here today with us to discuss these questions.  David Montgomery on my immediate right is a Vice President of CRA International, a consulting firm here in Washington where he directs CRA’s environmental practice.  David is an internationally recognized authority in energy and environmental policy, in regulation and particularly in questions related to modeling the effects of possible climate policies.  He previously held senior positions at the U.S. Department of Energy and the Congressional Budget Office.  David is one of the principal authors of an important new study of California’s climate law which was commissioned by EPRI, the Electrical Power Research Institute, and we are privileged to have David join us here today to present this study to a Washington audience for the first time.  This study uses a state-of-the-art modeling system to assess the impacts of California’s law under various implementation scenarios.

Dallas Burtraw, on the far end of the panel here, is a senior fellow at Resources for the Future based here in Washington and a member of the California Market Advisory Committee.  I should explain I’m introducing our speakers in the order in which they will present.  I’m not skipping Lee in the middle out of any slight.

Dallas is an expert in the design cost and benefits of environmental regulations and the regulation and the restructuring of the electricity industry.  Over the years he has taught at the John Hopkins School for Advanced International Studies, Georgetown University, and the University of Michigan.  I expect that Dallas will provide us a spirited critique of Mr. Montgomery’s presentation, and I’m sure everyone here is looking forward to hearing from him.

And finally, in the middle, Lee Lane is a consultant to CRA International where he works with David.  Before that he was the executive director of the Climate Policy Center, a Washington, D.C. based policy research organization that analyzes climate policy options and, at least, under his leadership worked to promote economically-efficient policy responses to the challenge of climate change.  I’m also proud to say he is the author of Strategic Options for Bush Administration Climate Policy, which we were privileged to publish here at the AEI Press in November, and I recommend that book to all of you, if you are not familiar with it.  I believe Lee is going to let David and Dallas duke it out over some of the specifics of David’s study, and instead will try to place this discussion of California’s climate law in a broader political and policy context that should be very interesting.  I think Lee and I share a growing suspicion that we are quite likely to embark upon a, well, I’ll simply say a disastrously ineffective approach to climate change.  And I think his remarks today will shed some light on why that is likely to be the case.

As is our usual format here, I have asked each speaker to talk for about 15 to 20 minutes and after that, after they conclude, I think we will have a little bit of an exchange of ideas between the speakers and follow up comments, and then I’ll open up the floor to all of you for your questions.

So thank you very much for coming.  This is a great turn out and without further ado.

David Montgomery:  [Speaking away from the microphone]

Samuel Thernstrom:  If you are going to stand, I think --

W. David Montgomery:  I’ll go behind the podium.

Samuel Thernstrom: Oh, okay, great, sure.  I just want you on the mic that is all.

W. David Montgomery:  I’ll start here.  I think the clicker here will work.  We will see [audio skips from 6.58 to 7.38] conclusions.  Thanks.  I’m going to have to ask you to click when it appears.  This is a little too far and actually I can give this back to you.  And that is the little sign that says ‘click’.  Thanks.

In doing this study we tried to address two kinds of questions.  The first was about California’s policies overall.  Are they likely to impose a cost; is meeting the targets that California set for itself likely to impose a cost on the state economy?  That is a question that is of interest to some extent in California, though the decision has already been made there about setting the targets themselves.  It is of interest to other states that may be thinking about adopting similar policies and, certainly, I think it interests in thinking about whether California serves as a model for broader federal legislation.

It may appear to be an obvious question, especially addressed to AEI, but immediately after the announcement of the government’s targets, at least three studies came out in California claiming that there would be substantial benefits to the California economy from these large interventions in energy markets and reductions in emissions, so even this is not an uncontroversial part of the study.  But what we really tried to focus on was the implementation of the California targets because once those targets were written into law, all of the important decisions actually still remain.  Dallas may disagree with this but my view is that the targets left just about everything that mattered up to California PA and Air Resources Board to decide at some point over the next 5 to 10 years.

So the other focus of the study was on how can these laws be implemented in a way that will avoid unnecessary costs and unanticipated effects on the economy and markets?  Now in order to answer that second question, we needed to develop a much broader and more capable set of analytical tools because the questions that come up, especially when we look at the kinds of policies that are being proposed in California, require a much deeper look at how in particular electricity markets work than we typically get out of the kind of economic models that are used for looking at state level policies or they were available in California at the time that these targets were announced.

So that is what we developed in the course of this study, but even though we developed some models - and I’ll talk about some models - what we tried to do in the study was actually concentrate on the issues and use the models to get insights into those issues rather than to produce numbers for their own sake that could be batted around about how the high costs work.

And the first question there was how large are the economic stakes in getting the decisions right?  Is it pretty much a matter of indifference how you do this?  Or are there really substantial stakes involved in how California decides to implement these targets?  Then the second, obviously following this, what are the key design choices?  What is it most important that California tries to get right?  And the third one, which economists love to look at, are there any potential unintended consequences or perverse results for energy markets that we might see out of the kind of choices that are currently being considered?

Dallas, if I could -- next click.  I’ll try to summarize quickly the principal findings, that way if I do not get to the end you will at least have something to ask me some more questions about.

The first conclusion is yes.  Greenhouse gas controls will impose cost on California in all the scenarios that we looked at.  The second point is that if we look at broad market-based policies such as cap-and-trade, they beat command-and-control regulations hands down.  We tried to look at this in several ways and actually in many ways it is the most difficult thing to model and what is most difficult to -- it is difficult in several ways.  In my opinion the most difficult part is that we tend to be too generous to the command-and-control regulations in just about any kind of economic modeling that we do.  And I’ll talk about the reasons for that. 

So we looked at how costs are increased if regulations are designed sector-by-sector without an effort to equalize marginal costs across sectors.  And we looked at what happens if you lay regulatory programs like fuel economy standards on top of the cap-and-trade system, which would already motivate changes in the area that those regulations are applying to.

The second point is uncertainty about future policies and goals matters, now.  In particular, in the electric power industry, essentially power plants last forever, and so, even policies 20 years out can have a significant effect on the choices that we have made over the next 5 to 10 years of what kind of generating capacity to build.  Not knowing what those policies are going to be can therefore impose costs today, both in terms of mistakes and also in an unwillingness to make capital investments until you know how that policy is going to turn out.  If there is also a short run risk in this area from the kind of schedule that AB 32 set for making decisions about how to implement the carbon caps in California, which has some disturbing similarities to the schedule on which California decided how to restructure its electric power industry in the 1990’s which contributed or were the main cause of the lack of generating capacity that produced the electricity crisis by the time we got to 2000.

The third point, and we may or may not have time to debate this today, is that when we looked out to the goal of reducing emissions to 80 percent below 1990 levels by 2050, what we saw from our modeling results was that that is essentially economically infeasible without technologies that simply are not in the cards today and that will need to be developed.

There were some other issues about designing a cap-and-trade system that we tried to look at.  One of them is the upstream or downstream issue.  Should a cap-and-trade system be imposed on the emissions of large emitters we can actually measure or should you back upstream and try to cap the carbon content of fuel so that you can capture, for example, all the motor vehicles in California by restricting the amount of carbon that may be delivered to them in gasoline?

Second, we were looking at offsets, that is, the ability to make money from selling credits on emission reductions outside the covered sectors:  forestry, agriculture, other greenhouse gases.  That is important and it can make a big difference in costs.

And finally, linkage to other systems.  There have been proposed -- part of what California was thinking was potentially being part of a larger western system that could deal with some of the difficulties California is facing in what do about electricity imports.  They were also talking about linking with the REGIS [sounds like] system and if they did that I think we might see some very unexpected results.  And so those may be some things we can talk about.

And finally, electricity.  I’ll spend a lot more time on electricity in the presentation than this last bullet indicates, but California also took on the task of regulating the emissions associated with imported power.  Even the state of California does not have extra territorial jurisdiction and that is very difficult to do.  We essentially could not find any way of California’s achieving that unilaterally that did not have some form of an undesirable consequence.

Let’s go to the next slide.  Now the history of this development was the governor issued an executive order in 2005 setting targets of reducing emissions to 1990 levels by 2020, and 80 percent below 1990 levels by 2050.  Assembly Bill 32 wrote the 2020 target into law but essentially left all the decisions up to the Air Resources Board about how to do it.  SB 1368 gave the California Public Utilities Commission the authority to regulate emissions from electric power starting today and also set the stage for the approach to emissions associated with imports of electricity.  There was a lot in AB 32 that seemed to push against the use of a broad, comprehensive market-based system and the governor responded to that with Executive Order S2006, which created the Market Advisory Committee and irritated some of the legislators in its push for a market-based system.

Let’s go on to the next slide.  To do this study, to say that in order to really get at the policy design issues in California, we needed a model that could provide insights into changes, behavior and market responses.  This both included broad design issues like cap-and-trade versus regulatory command-and-control, and then the details of how the power sector will be regulated. 

So we took our broad economy of modeled U.S. economy which we call MRN and our very detailed model like the power sector that we call NIM [sounds like] and we actually merged the two of them together so that we could do both of these things with the same model.  And try, therefore, to be able to actually understand the incentives created by policies by looking at how rational economic agents within the model responded to them.  And this, I think, gave us the kind of modeling capabilities that we required.  It is a model that is based on sound economic principles: supply and demand work, and individuals and businesses make rational decisions.  It had enough detail that we could figure out what was going on in the electric power sector and tell the difference between the complicated rules that the PUC is coming up for how to measure emissions associated with imports.  We have long enough time horizons that our agents were not going to be fooled about what was happening in the future.  We could have that as part of our conditions for consistency that their expectations matched with what actually is going to happen and enough detail in regional and sectoral impacts actually see how California’s trade with the rest of the U.S. is affected.

So now let’s look at some results.  There will be a cost.  This is the range of impacts on California Gross State Product.  We come up with a -- from all scenarios that meet the targets through 2020, they differ in some cases in optimism or pessimism about underlying factors, in some cases about policy design and in some cases about what we assumed the targets were going out from 2030 to 2050.  And you can see the impacts at their peak range from about a half a percent of GDP to about 3 percent of GDP for the state.

Carbon prices depend on the policy outlook and the design of the policies, and it looks to us as if it is going to be tough for California to meet these limits.  And that is largely because so much of California’s emissions come from transportation, which is incredibly slow and costly to change.  The electric power sector, therefore, will take on a very large part of the burden, but even if you drive electric power emissions down to zero, it is hard to get much beyond that 2020 target.  And so that takes really higher carbon prices to do that, ranging from something in the US$10 to US$40 range in 2015, getting up to US$100 a ton potentially by 2020.  In the high cases, interestingly having a -- if the market was confident that they were going to be tight targets in 2050, our model predicts you would start to see some investing today in things like development of alternative fuels for vehicles where there is learning by doing because you need to invest today in order to build the capacity to serve the market in the future.  So we get the results like this but US$100 a ton of carbon is approximately a dollar a gallon of gasoline, and approximately US$100 per megawatt hour of electricity.  So that is a big number.

On the next slide, price volatility.  We want to be able to talk about the safety valve issue today, so I want to be sure to get this one in.  California is imposing hard caps.  There is no safety valve or escape clause except the governor’s declaring an emergency and therefore delaying the cap by one year.  There are a lot of things that can change the cost of meeting a hard cap.  Two of them that matter if you are looking at electricity are natural gas prices and electricity demand.  If electricity demand spikes in a year because of weather or if natural gas prices spike you can get -- it can become very much more expensive to meet an emission cap than if they do not.

So what we did was we compared in this case -- first we took our general midcase that we called our peer trade case.  And we just varied up and down electricity growth and natural gas prices from one year to the other.  And that gives us sort of that purplish spiking and cyclical behavior of carbon prices.  We compare it to the kind of price that we would have had in exactly that same case with exactly the same assumptions if we had hit the averages for gas and electricity prices in every year rather than having them bumping up and down.  And then if you look at the harder to see -- and that shows us what happens to prices, lots of volatility in prices.  When we look at the emissions, we see it makes almost no difference to emissions whether you allow this volatility or you prevent it; and certainly, it averages out over time.  Therefore, this lack of a safety valve produces the prospects of completely unnecessary price volatility in California and yet it seems to be something that was ruled out from the very beginning - let’s go to the next slide - in the design of the policies.

We actually tried to look at another view of the safety valve which is California got into these policies after having heard from a lot of economists at the University of California and in state agencies that it was not going to cost much at all.  Well, if California got into this thinking it was not going to cost much at all and expected carbon prices to be pretty low and the economic impacts to be pretty low, and then it turns out they are not, we are in for a really unpleasant surprise. 

So what we did was we took the optimistic assumptions that we thought kind of corresponded to what the state has originally -- some of these original studies have been thinking.  We put them into our model and saw what it would cost -- what the carbon price would be if meeting California’s carbon caps with those optimistic assumptions.  Then we said, “Suppose they decide to put their money where their mouth is and say okay, we believe that it is not going to cost much so we are going to put a cap -- so we are just going to set the carbon price.”  And this, in fact, is if you like a two-way safety valve.  So we just said, “Let’s set the carbon price and see what the costs turn out to be, if the assumptions are not correct.”  Well, having the safety valve reduced the cost of our peer trade case by about 50 percent.  It gave you less emission reductions, but it kept the costs to what you would have expected them to be pretty much had the optimistic assumptions been true.

Let’s go on to the next slide.  In here we illustrate a lot of things about how policy design matters.  The red line connects targets of differing intensities so the safety valve is what we call SDLCA that gets less emission reductions in our peer trade case, which is hitting the 2020 target and then staying there forever.  Or our 40 percent case, where you ratchet it down to 40 percent below 1990 levels by 2050 or our 80 percent case, which gets you down to 80 percent below 1990.  We have plotted the change in welfare for the state of California, an overall indicator of economic consequences against cumulative emissions under these policies, and as you expect you get what you pay for and their increasing marginal costs.  The more you want to try for, the faster the costs are going to go up.

The interesting part is the range around the peer trade case where we found that either optimistic assumptions or availability of low-cost offsets or the presence of market failures that you could actually fix through you policies or allowing a lot of leakage in the electric power sector; all can reduce costs a lot.  And on the other side, poorly designed regulatory programs, which we represented through sector specific caps or mandatory efficiency standards that go far beyond what individuals would rationally choose to do on their own both increase the costs substantially.

Let’s go to the next slide.  And this is just a bit of thinking that I would like to go through about regulatory policies.  As I said to begin with, I think that we seriously underestimate the cost of regulatory policies in our models because the power of the regulatory policy is it can never search out all the decisions that individuals can make and get them all right.  Well, neither can a model.  And there is a problem of eating the menu when you are a modeler which is to say, “We have represented decisions in our model, and we have done something with the model to show what happens, if you impose regulations, but we can never capture all the ways that a regulator can get it wrong because we cannot capture all of the ways that individuals can get it right by doing something different from what the regulator is telling them to do.”

Let’s move on to one -- let’s skip the next slide, Dallas.  Let’s skip the next slide; although, it is extremely interesting.  And go to the point about trying to regulate electricity.  California is trying to regulate the emissions associated with imported power and it has defined a way of doing that in the rules that the California PUC has issued under SP1368.  And essentially, what the rules do is they say there is a standard for the emissions from the power purchased by the utilities in California.  The utilities that serve customers in California do not generate power; they just buy it.  They buy some of it instate some of it out of state.  The standard is on average the emissions from the power they buy cannot be any higher than what you would get from a medium good gas fired generator.  That is a lot less than you get from a coal fire generator and right now California is importing a lot of coal fire generation.

Well, how do you measure the emissions associated with imports?  What California said was if you have a long term contract with a specific unit that -- actually what they said was for making long term purchase of electricity, you must contract with a specific unit so that we can track what the emissions are from that unit.  But if you buy in the short-term market, we are just going to attribute the average emission factor. 

Well, that is where we start getting some perverse consequences.  The result of these rules is likely to be either increased costs in the state of California or a tremendous amount of leakage in which attempts to reduce emissions in California are all completely offset by increases in emissions outside the state of California.  The way you would get leakage is by something that has become called contract shuffling.  Basically, California right now contracts for some of the power that is produced in the West but by no means all of it.  A lot of the power produced in the West is hydro, wind and nuclear that is used by people other than Californians.  The simplest way to meet the CPUC’s rules is simply to sell off all your interests in coal fire power plants and buy up all those existing contracts from existing units.  It relieves the pressure to reduce emissions within the state of California, so you get more emissions in California than you would with an alternative rule; and exactly the same thing is going on outside the state.

The problem is that if this happens it could -- there are several consequences.  The first one is there is just going to be a lot of contract shuffling going on out there that is going to be opportunities for traders and investors and brokers to try to figure out where to make money.  The second is if that contract shuffling takes place, there is going to be -- California is going to have failed to succeed in a lot of its goals.  The third one is, there is an alternative, which is to wash the emissions from coal fire power plants by selling in the short-term market rather than the long-term market.  That is a really chilling thought because if the long-term market dries up for California, if the utilities start buying again in the short-term market because of these rules that is 2000 all over again because that is what the regulatory program of the ‘90s told the utilities they had to do; and so, they were completely exposed to all the vagaries of the Western natural gas market and electric capacity markets with the result that this could produce much more volatile prices in the West and affect investors in other parts of the region, as well as California.  So depending on how California does this, leaning against contract shuffling can produce very high costs for California by restricting the choices of utilities.  Not leaning against contract shuffling is going to defeat many of the purposes of the legislation.

Let me use those to kind of tee-up some later discussion and I’ll stop for now.  Thanks.

Samuel Thernstrom:  Thanks very much.

W. David Montgomery:  Okay.

Dallas Burtraw:  How about if I just take it from here or I can do the clicker.  Thank you David and thank you everyone for being here and thank you for the invitation to be here.  I’m going to offer the usual disclaimer here with some emphasis; I’m not speaking in behalf of the Governor of the state of California or any members of the Market Advisory Committee.  To be sure, we were not always of the same mind anyway.  These are my views and I want to give you some perspectives on David’s -- on the report that David has summarized in his own comments and some other context for what is going on in California right now.

So the issue in California -- there are really two types of people in California, and I’m from California so I can speak freely.  There are people in California that feel that the most important thing to do is to get it right and that they cannot control what happens outside of California, what happens to the rest of the emission in the world, but they are going to get their own house in order.  And then there is another school of thought, which is that their goal is to really influence what is happening on the national and international stage.  And on the Market Advisory Committee that I’ll give you a little bit of insight on the results that are coming out of that.  There is an interplay here of both these things.

A key issue is this questionable leakage which David has set up and it is much larger than just on the electricity sector because it is a possibility that cement or petroleum which -- there is relatively dirty petroleum in the state resources that could be sent out of state for refining if certain types of restrictions are put on refining instate and not cause an incentive to move production out of state.

So leakage could be as broad it is just that overall pattern of economic activity in California and in the Western states, but there is a good chance that California will not act alone.  Currently there is a Western state initiative, including six Western states, Washington, Oregon, California, Arizona and New Mexico along with Utah, which most recently joined and two Canadian provinces and they are meeting on a monthly basis the same kind of relationship and format and with some of the same staff supporting that led to a 2-year process that is part of the Northeast Regional Greenhouse Gas initiative. 

So in what is in David’s report, it is in some sense a worst case scenario because it is California going it alone, and in my mind that is off the equilibrium path outcome and that California cannot go it alone all the way anyway.  But what David’s report is telling us is, well, what would happen if they were to try to go that far using the tools it has available to it.  And of course, California is developing some architecture that I think contributes importantly on the national stage.

So here are some key features that are in our Market Advisory Committee recommendations.  These are the draft recommendations from the end of last month which was delivered on time, and you will have on Saturday on time the final recommendations of the Market Advisory Committee.  And here is a couple of the most -- this is a 14-member committee and 10 of those members are actually from outside of California and three of them are from Europe on this committee.  We are recommending an economy-wide approach to be achieved as soon as possible.  There is a recognition that monitoring data may not be available and some administrative capability may not exist but for recovering what would be roughly 83 or 84 percent of the state’s emissions excluding most importantly biological emissions; it is thought that within a very few short years that the infrastructure could be put in place to include the entire state under an economy-wide approach.

Second, there is adoption of the principle to move to a 100 percent auction as soon as possible.  It is not argued that the state start out at a 100 percent auction, but instead that a mixed approach be used in which an important fraction of the allowances are auctioned initially and some are distributed for free.  But the principle that is guiding free distribution here is not grandfathering [sounds like] as is the case in say the SO2 trading program which is for free to incumbent producers and emitters based on some historic metric but rather that it really is -- there are two things that are guiding it.  One is that it softens the blow on consumer impacts so there is a discussion of free allocation to [indiscernible] to a trustee on behalf of consumers that would be through the load serving entities that would soften the impact on electricity prices during a transition period, and that that of course could be phased out.  And there may be a call for some targeted free allocation to a couple of specifically vulnerable sectors including cement, which represents really a very small fraction, but that allocation would be contingent on continued economic operation of the state that would be an output-based type of an allocation.

The committee also mentioned or provided the preference that the state move towards the allocation of allowance auction revenues rather than the allowances themselves in order to avoid the perceived entitlement to free emission allowances for some time into a definite future and instead that the allowances be liquidated in the market and then revenues be designated to these purposes as much as possible.  There is no safety valve and there are generous offsets, which do help serve to contain costs.

Now I want to talk for a moment about electricity sector issues, which David teed up, and they are just really important.  Why are they really important?  Well, nationally the electricity sector represents about 40 percent of our CO2 emissions but it is expected to do about 70 or 75 percent of the emission reductions that we would see over the first decade or two of a moderate or fairly stringent climate policy.  In California, electricity sector emissions are a little more than 20 percent of emissions in the state but half of those emissions associated with electricity come from out of state.  The state imports about 20 percent of its electricity, but that stands for about half of the greenhouse gas emissions that are associated with electricity consumption.  And since, nationally and in previous programs, the electricity sector has played such a prominent role that it seems particularly important that California get it right in the electricity sector if it is going to try to do an economy-wide strategy at all. 

So the approach that is -- I do not want to go too far down this path because it could be time consuming -- but the approach that is being suggested is a first-seller approach in lieu of a load-based approach, which the PUC had gone down that path for a couple of years.  A load-based approach would have essentially identified the load serving entities as being the responsible compliant party.  And a first-seller approach is fairly analogous to things that we have seen before which would be a source-based approach.  The first-seller is identified as the party or entity that is first selling power into the California electricity system.  For instate sources this would be the power plant owner or operator themselves and for out of state emissions this would be the contractual parties identified on NERC E-tag documents.  These are electronic documents that have to be associated with every power transfer between control areas, so there is an identified party that is bringing power into the state and putting it into the grid, and that party would have a compliance obligation to them.

Some of the criteria that we used to compare these are the important role that the state agencies already have in portfolio planning.  The promotion that the state agencies are doing, which expect to end-use efficiency, the opportunity cushion, the impacts on consumers and for all these things actually there is not a distinction to be made here between both -- depending on how you set up the architecture a first-seller approach or a load-based approach can accomplish these criteria, like them or not, to an important degree.

Then there is this question of leakage and the EPRI study that David has talked about and the Bushnell et al study at the University of California has suggested that there could be 100 percent leakage. In other words, more than just the electricity sector reductions could be made up through increase in emissions out of state in shuffling of contracts out of state.  But I think that there is much more reason for encouragement than this kind of result that falls out of the modeling exercise because of the difficulty of such modeling to actually capture the institutions that are in place.  So we point to, for example, the California Energy Commission study that used the California climate action registry protocol to assign emissions at the borders essentially to be identified with the generation that is leading to the parties that are identified on these E-tag documents, and they find that 56 percent of the power can be uniquely assigned to power plants out of state; that is 56 percent of the power that is coming in.  And then of the remaining 44 percent that is unassigned they conclude that 96 percent of this is natural gas and only 4 percent is associated with coal out of state, which leaves very little room in fact, at least, in the short and the medium term for contract shuffling.

Now there still is a possibility of moving from long-term contracts to short-term contracts, like David mentioned, but the procurement standard that PUC is -- the way they are implementing it is they are monitoring contracts and they will disallow some of these contract shuffling and that if there is a long-term contract in place they are supposedly going to assume the continuation of course of that contract and not allow the utilities to slip out from under that contract.  But what happens if they do slip out of that contract?  It still has potentially a profound effect on the investment climate in the Western states and that is what you would read if you looked at the trade press, that there are a lot of parties and investors in Wyoming and elsewhere who have said, “Well yeah, this has an effect in terms of what the expected prognosis for them on profitability in the long run than in building new uncontrolled coal fire power plants in the west.”  So I think it has affected the investment climate; that all that in place, it still is a dangerous issue, this question of leakage.

And then there are some other issues with respect to why a first-seller approach really dominates a load-based approach.  Now there is an especially sophisticated new market design in California, which is going to reintroduce a day-ahead market in 2008.  The first-seller approach is absolutely critical to overcoming that, and the load-based approach would essentially remove all incentives and ability of the load serving entities to express preferences upstream to the generators.

And finally I’ll just say that accountability and transparency, I think that a key component of the success of the SO2 trading program has been virtually 100 percent certainty of emissions monitoring and you saw it within a few short years, environmental advocates who were very much hostile towards cap-and-trade type approaches becoming the biggest advocates, and I think that was because these programs were so -- there was so much confidence that the continuous emissions monitoring for the SO2 program, for example, was working and working precisely.  And so any kind of approximation method that might be associated with the program in general or a load-based approach in particular I think is lethal to establishing public confidence or for building a market that can inspire investment confidence.

Well, David’s talk is addressing what matters most than what is missing and using a modeling tool that -- I mean I’m a modeler and I have used both kinds of tools, a CG type of framework and the more detailed sector models - and I do not have a horse in this race - so I can just sort of be on the sidelines and express the view that all models provide important insights and that applies here.  But I do want to put on the table what I think are the limitations of this modeling framework, and it is driven in large part by the question that had to be asked, and any model would fail in this regard, and that is looking at a 50-year time horizon. 

I find it a lot more interesting what is suggested between now and 2020 than what is suggested between now and 2050.  And even by 2020, it is difficult to know what are going to be the changes in technologies and the changes in preferences that are going to affect how well and how easily California can come into compliance.  And you know when you are talking about change in technologies we are not just talking about carbon caps and storage for coal plants.  We are talking about end-use sufficiency technologies that could dramatically change the cost of lighting and cooling and heating in buildings, et cetera.  And when we are talking about change in preferences, I just had to reflect as I was walking over today in this tie, if you had gone into a southern utility 15 years ago you would have found everybody dressed like this, and today you go down there and you are not going to find any ties.  You are going to find polo shirts, and that was a change in preferences associated with end-use -- reduced the energy consumption and end-use and it was not painful, as a matter of fact lethal for the people that have swapped out of the tie.  So changes in preferences can also matter importantly.

Okay, I’m going to watch the clock and try to wrap up in just five more minutes. 

The policy targets that are identified in AB 32 were above our pay grade and for sure the costs should matter, so the important lesson that is in EPRI and Charles River study is that the costs matter in terms of studying the policy goal, but we also need to wake up to the emerging social norm that appears to be that a commitment to achieving climate goals and this is showing up in poll after poll these days.

And I do not want to take us to the point of saying - at least for the purposes of this conversation - that clean air is a right or that protecting the climate as we know it today is a right, although some in this room may believe that.  But I will say that if you really look at something that is rights-based like the Civil War there would have been, to end slavery, people would have incurred any costs.  And we have an emerging social norm that many in our society are adopting the view that they will incur any cost to arrest the rapidly changing climate.  And economists still have a way to contribute in this discussion because if we could have avoided the bloodshed of the Civil War through some more insightful solution - that would have been great.  And to the extent that the modeling can tell us that there is one instrument or tool that is available to us that can help us achieve these goals than the other then that would be really great.

One of the issues that comes up than is, to extend the metaphor of the Civil War, cap-and-trade versus regulation and we had several public hearings in California, and I started to see a pattern emerge and here again there were two kinds of things we heard.  There are those that said, “You have got to do everything you can with market-based approaches.  You get everything into the market and then whatever is left over and you cannot quite get covered in them inside the market and you sweep it into some kind of technology program or regulatory program.” 

And then there were the others that said, “Well, you have got to do everything you can with regulatory programs so we get these emission reductions for certain by date certain and for sure and we can monitor and enforce these technological changes; and then, well, whatever is left over then maybe we will sweep it into a cap-and-trade market-based program.”  And this really distinguishes the debate that is going on in the California legislature right now.  And I think the insight that is in our recommendations is that it is not an either/or proposition; it is especially not since many of the technology programs are in fact a done deal.  They are going to be in place and then the question was what kind of cap-and-trade program helped to achieve climate goals even more cheaply given that there is going to be many technology measures in place?

And -- let me skip that and just go on -- here are some of the technology measures that are going to be in place.  The low carbon fuel standard, which I do not think is represented because your study got initiated before that came to being, right?

W. David Montgomery:  True but we actually are actually back stopped -- is a low-carbon fuel and we get a lot of it so I think that we may end up meeting it.

Dallas:  It may be represented, okay.  And the low-carbon fuel standard, though, is a well-to-wheels standard so it is looking at changes from the refining process and extraction, even all the way through to the use of fuel in a vehicle, and it allows for all sorts of fuels to compete and including electric vehicles and hybrids in some context that has to be thought through.  There is the Pavley Bill, which is the tilt, so low-carbon fuel standard is targeting fuels in transportation, the Pavley Bill is targeting tailpipe standard; and so essentially, the engineering hardware that is involved in transportation. 

We have that Senate Bill 1368, which David mentioned, which is the procurement standard enforcing a greenhouse gas standard for energy purchases on long-term contracts.  And then a whole sweep, a growing sweep of investments in energy efficiency including decoupling of revenues from sales to try to remove the disincentive for utilities to invest in energy efficiency programs and then taking things one step further, that proactive incentive by adding allowing shareholders to profit from investments in end-use sufficiency by putting into the rate-based and earning rates of return on hardware that is put in place.  And the decision on this latter measure is expected from the PUC this fall.

So I think I’ll go through the next two slides and then I better stop.  So the question can incentive-based policies like cap-and-trade do the whole job?  And I guess I’ll just show you my -- I’m pluralistic on this.  I think that incentive-based policies are essential but no, I do not think that they can do the whole job and there are many critiques of efficient markets in specific context.  So you could ask:  Do cap-and-trade programs when you go to ubiquitous pollutants such as CO2, is there death by a thousand lashes? 

Essentially, there is a lot of small little bleeds (sounds like) where certain elements of the residential sector where time of purchase -- home owners - the last thing they are considering is energy efficiency of a home or separation of metering for apartment units, et cetera -- and you start adding these things up and all of them individually sound pretty insignificant, but many would allege that taken into whole, you have the problem that a market-based approach in the use of prices -- the prices would have to rise to such a substantial level, essentially, in order to reach all these and cause institutional changes to be able to reach all these that may be politically unsustainable.  That is the justification for the use of technology polices and an example is discount rates and terms of the kinds of choices that consumers make. 

I just picked this up because it is one that advocates will most often site.  And you can ask, well is this consumer irrationality or is it a test of fitness?  And those consumers that use stupid discount rates when deciding what kind refrigerator to buy, well, they are going to end up with less disposable income than everyone else; but hopefully, there is going to be learning over time.  But the fact is that this is a pervasive factor that influences consumer decisions.  And so, should policy try to address this through appliance standards and building standards and things like that?  But it is not all those dumb consumers that we are talking about.  These kinds of issues are also ubiquitous in business culture and the one that comes to mind is regulatory institutions for transmissions.

So I think it is interesting to draw on the ideas of a former president of EPRI or director-president-CEO of EPRI who is fond of pointing out, and EPRI and previous studies have pointed, that there are US$50 billion in annual costs associated with congestion on the transmission grid that could be remedied with investments that have a one year payback, yet this low hanging fruit appears to be unpicked, year after year, after year; and EPRI has been -- after doing research they try to identify why is this low hanging fruit not getting picked and they have been talking to the Congress about the fact that certain institutional reforms have to happen in order to allow this to be harvested.  But this continues to persist. So whether there are US$20 bills on the ground or not, I think that there is low hanging fruit that is available to us.  And sometimes you need direct policies in order to be able to reach this.

And then finally, there are the limits of political support of price in cement so they are just -- and last week was in the news -- you may have seen the Stanford University RFF survey in which the U.S. public is showing a clear preference for action in the electricity sector over their own transportation behavior, and secondly, a preference for standards over cap-and-trade or taxes.  And one thing that is in this report is that -- David skipped over that slide - but maybe I’ll invite him to come back to it - about RPS.  One of the things about renewable portfolio standards is they are far less efficient at achieving carbon emissions than would as a direct carbon tax.  But one thing about them, they tend to keep electricity prices low and it tends to be producer surplus that gets eaten rather than consumer surplus through these kinds of programs; and some really with the low-carbon fuel center, I think that is the way it takes effect in the transportation sector.  So consumers and voters may not be entirely irrational in terms of the kinds of preferences they see and they express for different types of policies.

So I think that the role of cap-and-trade with respect to vis-à-vis widespread technology programs is I prefer this viewpoint:  Leave no low-cost emission reductions behind.  And that is the role for cap-and-trade going forward, especially in California, given that the sweep of technology policies that are going to be in place are needed anyway, but those policies, by themselves are also destined to fail because it is just a shock-end approach that is going to miss so much and fail to put through price signals that are going to affect people’s behavior.  So I think that these are complementary approaches and I think that is what is going to be needed in order to achieve climate goals.

Let me stop there.

Samuel Thernstrom:  Thanks very much, Dallas.  That was great.  Lee?

Lee Lane:  It is obvious, I think, that when you compare the AV equipment of the people on the panel that I’m a Central European historian talking at a panel with two economists; and in fact, I’m going to somewhat shift the direction of the discussion, at least for the last of my remarks and talk a little bit about the politics of greenhouse gas emissions controls rather than talking about the economics.  Then we can revisit the economic issues in the longer discussion.

In preparing for this session, I actually partly by chance, took off my shelf an old copy that I had of David Mayhew’s Congress: The Electoral Connection, which is a book that is now about 30 years old, and I ended up liking it so much that I reread it from cover to cover again.  And one of the things that Mayhew says early on in the book, I wonder how my colleagues would respond to this, is that what we really need, and he is talking specifically from the standpoint of political science, but “What we really need,” he said, “is to somehow revive the 19th century discipline of political economy.”  Perhaps to a considerable degree, at least, in some fields the subsequent 30 years actually did a bit of that, but my thesis here this afternoon - a part of my thesis here this afternoon - that the area of climate policy in particular would benefit by a sharper focus on the intersection and the interaction of political and economic analysis and that that interaction is really a place that might generate some insights that are currently not as prominent in the discussion of climate policy options as they might ideally be.

When I look at the current larger scheme of climate policy, and it is sort of starting with California, but not necessarily limited to just California, it seems to me, as Sam suggested, that a lot of what is happening seems to have the nation and the world pointed towards a pretty inefficient outcome.  We certainly seem to be headed toward geographic patchwork of emissions limitations that miss some of the most important countries, some of the most important sources of emissions.  To a considerable degree, I think both David and Dallas point out that this is not absolutely clear; but nevertheless, to a considerable degree I think we are pointed toward quantity-based controls rather than price-based controls; whereas, most economists I think would tend to suggest that with a pollutant like greenhouse gases, we would probably be better off with a price-based system.  We are certainly moving rapidly not just in California, but clearly in the U.S. Congress, toward command and control systems layered on top of or put in place of even the cap-and-trade systems.  And a lot of the policies going forward seem to me to be building a huge amount of regulatory uncertainty into the future. 

The tremendous uncertainties that David’s presentation pointed to in California about whether we are really going to be aiming for 80 percent emissions reductions by 2050, something which I noticed Congressman Dingle announced just yesterday might be in the new House bill.  To me, that is just a clear signal that nobody is going to know for sure what the real targets for 2050 will be for a long time. 

So all of those policy trends are, I think, at least -- and I hope Dallas would agree with me -- at least potential sources of inefficiency in a way that the system is developing; and they are not, by no means, just in California.  I think you see the same pattern in the legislative proposals at the federal level in the U.S.  It happened a couple of years ago that I spent a good deal of time studying German climate policies that existed then, you certainly see the same patterns there.  I suspect they exist in several of the European countries, as well.

So if there is a pattern, it seems to me and it is a pattern that perhaps suggest the existence of some serious potential problems, we ought to ask ourselves why this pattern?  Perhaps part of the answer to why is evident on the face of the matter; wherever one finds these greenhouse gas emissions control, policies being put in place, a common element is a strong advocacy by some of the environmental advocacy groups.  And it is not surprising that those groups cross a lot of geographical areas, have very similar attitudes and objectives.

I have talked a lot to representatives of those organizations and they are usually pretty candid about saying that economic efficiency is not their organization’s real priority.  So it is not very surprising that the proposals that they make do not place a great deal of stress on economic efficiency as one of the leading values.  In fact, the only thing that it seems to me would be likely to change that outcome would be if the political market, that is to say if the legislative assemblies to which they are trying to sell these proposals itself, demanded highly cost-effective policies.  Well, the political science literature of this is very, very clear cut that the very opposite is true; that those markets are in fact not at all driven by economic efficiency or cost effectiveness considerations and indeed are systematically biased against the kind of environmental policies that are likely to produce economic efficiency.

There are a number, I think, of interesting examples that I could probably spend more time than we have talking about it.  There are so many examples, but let me just very quickly list a couple of them. 

Dallas referred to the objective in California of using the imposition of emissions controls in California as a stimulant for encouraging first the United States and then the world to reduce greenhouse gas emissions.  And fundamentally, this boils down to an idea that California is going to establish a moral example that will induce the United States to adapt emissions controls, and that will establish a moral example and the result of that will be to persuade India, China, and the rest of the world to do the same thing.

Well, that is an interesting thesis and there is at least one school of thought in international relations that says that that might be conceivable, but there is also a great deal of analysis that says that just is not the way the world works.  And in fact, David Montgomery has done some analysis suggesting that unilaterally imposing emissions controls actually creates barriers or incentives against other countries adopting similar controls.  And Scott Bared [phonetic], coming from a completely different point of view, also argued that this just is not likely to work.  All right, so nevertheless we have this line of thought being extremely persuasive apparently it was with the California legislature, and it seems to be in the Congress, and yet there is a lot of at least cogent-looking analysis that says it is not really economically very logical.

Dallas alluded - let me pick a second example quickly - to the survey, and I thought it was a really interesting survey incidentally, suggesting basically that the public just does not like and approve of the use of price incentives as a tool of environmental policy.  Not only did this survey produce that result, but I have to say I have looked at this literature and there are really a number of surveys that all showed the same thing.  A new book by Bryan Caplan makes an argument that, in effect, not just the U.S. public but probably most electorates just do not understand and disapprove of the use of price incentives as ways of achieving social objectives.

Well, insofar as politicians try to please electorates, which after all is a basic concept of representative government, that suggests that there is a pretty strong incentive on the part of many representative governments against the varied kinds of policy tools that are most likely to be cost effective.  There is the really simple fact that there are really no institutions that are committed to selling the idea of economically efficient or cost-effective policies.  I mean, the environmental groups have their goals, the business has a fiduciary responsibility to its stockholders; but whose job is it to ensure that the policies that are enacted are economically efficient and cost-effective ones?  The answer is really nobody; nobody’s job.

In so far as anyone plays that role institutionally, in the past, it has been the executive branch of the government of the United States.  And because of the particular political constellation in which a climate policy has been made up to this point, the executive branch has not really engaged in the debate about the form of emissions control should they eventually be enacted.  And maybe one of the rays of hope that we might look for is that at some point if the executive branch becomes more involved in the process, it might somewhat change the political -- the equation.  Anyway, I think it is worth thinking about.

And then, finally, just one last example.  One of the things that Mayhew observed 30 years ago was that a very strong theme of legislative action was symbolism; that legislatures tended to enact things that looked as though they would have an effect that actually did not do very much or did much less than they appeared to do.  And it is not a totally bad thing if a symbolism is relatively cheap; it could be the best available policy option.  But I wonder if the proclamation of 80 percent emissions reductions by 2050 is in the category of symbolism.  And one of those things that legislators or others hope that they can say that will not impose immediate cost -- and after all as you get closer to 2050, there is always a chance to revise whatever measure it was that called for the potentially very expensive emissions reductions.  This race is another piece of political science literature, Shep Melnick’s piece about regulatory failure that I think is worth thinking about here.

Sometimes when I think about these observations, they seem so obvious to me, I wonder if I'm saying anything new or useful to anyone.  And yet, it does seem to me that the inefficient tendencies of government, the propensity to enact not cost-effective policies have real consequences for climate policies that I do not see many people talking about.  The first of those consequences is an obvious one in comparison with the modeling that I have seen about the cost of Kyoto and the cost of stabilization based on the assumption of a beautiful worldwide uniform carbon tax.  If real world policies are likely to cost a lot more than the ideal policies, that seems like a fairly important potential finding.

Secondly, there is a real chance they will fail; that is to say that the policies that we are working on to reduce greenhouse gas emissions may simply not be effective enough and adapted widely enough to actually avoid serious future harm from climate change.  I'm not at all claiming that climate change might not be quite a serious problem.  And if that is true, I mean, if there is a serious possibility, it would seem to me that the logical conclusion would be to look at alternative policies like adaptation and geo-engineering, and to apply a lot more emphasis on R&D in that direction than is currently the case, where, in fact, the government is spending almost nothing on policies other than the greenhouse gas controls.

The third consequence that seems to me to emerge from all this is that failed policies are sometimes reversed.  Not always, by any means, but sometimes they are and sometimes they are reversed quite rapidly.  If you look back to things like catastrophic Medicare coverage, for example, and there are several other examples, how stable are the policies that we are evolving toward? 

I think both David and Sam observed that it seems pretty likely they were going to end up with some version of cap-and-trade, and I think that is true.  But my suspicion is that there is a chance that the first version, anyway, of what we enact is going to be changed, and perhaps quite dramatically.  And one of the things we might think about doing as a way of looking forward to climate policy is thinking about how we encourage learning and changes in the policy that take advantage of the initial experiences.  Now, maybe all these things will happen automatically, but there are lots of ways of doing that.  It is not just things like sun-setting. 

I think actually an interesting aspect of a point that Dallas made, I wonder if the allowance allocation regime might not influence the rigidity of future cap-and-trade policies one way or the other.  It seems to me the more grandfathering there is, the more institutional inertia there may be behind whatever scheme is created.  It is a complex subject; I merely raise it.

So those would be my thoughts.  Sam and I are working on a new project and a new proposal to spend a lot more time and energy thinking about these things and researching them, and we hope we get a chance to do it.  I’ll be glad to talk about any of these ideas and I certainly welcome the other panelists’ thoughts on them as well.  Thank you.

Samuel Thernstrom:  Thanks very much, Lee.  That is great.  As I promised, I think I would like to give all the panelists now a chance to comment on each other’s presentations.  I saw David took quite a few notes while Dallas was speaking.  Maybe I’ll give you the first crack at the mic.

W. David Montgomery:  Thank you.  Let me start with just one.  I was sorting through them and I guess the image that was striking me was of some rich southern boy arriving at Harvard in 1860 and what he felt like the first time he met an abolitionist.  And I have had the same feeling in meeting these serious proponents of regulatory approaches and market failures.  But Dallas is not in that category because Dallas does identify something that is, in fact, a serious economic question: Are there market failures which are leading to less than efficient choices of how energy is used, and what kind of policies can be used to address those?  Certainly, if we could design policies that went directly at those market failures, there is a substantial chance of harvesting some of those low-cost emission reductions.

I’ll make two observations, but allow me to stick with just one observation on that.  The real problem is that nobody has actually tried to design policies that way.  That what we see are market failures being used to rationalize the same tired old central planning solutions that have been around for a very long time.  And it is not clear to me that -- and so that is the challenge for California, that there is a -- the policies actually have to be designed to deal with the market failure.  You cannot just say, “Oh, there is a market failure there so I'm going to go and tell everybody what to do.”  That is not the solution.

The second point is you cannot fix that solution by sticking a cap-and-trade program on as a kind of residual measure to sweep up what the regulatory programs have not missed.  Because if the regulatory programs are binding either in broad or in detail, what they are going to do is force out cost-effective measures for meeting that cap and force in something, which the regulatory measure requires be done.  So they raise costs.  It is not like the old regulation of sulfur emissions from electric power where we could allow emission trading in it fix many of the inefficiencies of the old system.  This cannot fix the inefficiencies -- emission trading cannot fix the inefficiencies of having regulatory programs there.  Those regulatory programs are just going to drive out the efficient things that the cap-and-trade program would bring about.

So, I would end up with a different slide than Dallas.  I would say the point of cap-and-trade is “allow no unnecessarily costly measures in.”

Samuel Thernstrom:  Dallas, any thoughts on that?

Dallas Burtraw:  Yeah, just a quick reply.  I appreciate what you are saying, David.

Market failures could be used to rationalize essential planning solutions and that, in any case, these technology policies are going to raise cost.  I agree, especially in the context of the modeling tools that we have available to us.  There is no question that the direct regulation has to be more -- by construction, it is more expensive than will the least cost solution, which is going to be a tax or potentially a cap-and-trade program.  I think that is true in the -- I know that -- we know that is true in the models.  I think it is largely true in the real world that these direct technology programs raise cost compared to the ideal price instrument or the ideal cap-and-trade instrument.  But there are these preexisting institutional barriers that make it so that sometimes the prices do not reach in as effectively as they need to, to some segments of the population.  So, what you would like to see is a sort of surgical behavior to fix the problems where there are real market failures.  I think if you look at the fleet of policies that are emerging in California, yeah, for sure the California agencies are technically among the most sophisticated agencies in the world, and that includes here in Washington.  Pound for pound, they have been very effective, maybe at a high cost, but they have accomplished the job that has been given to them over the last 30 years.

And so, if you look at AB 32, it is remarkably short.  It sets its goals and pretty much says, “Go do it.”  Some people’s initial response to that if you are outside California is like, “This is not anything.  Do not take that -- that is not the right interpretation for the way,” business gets done in California because in California these agencies have a legacy and have been given tremendous authority to achieve broad-sweeping goals, and they have done it; but many economists would look at it and said, “Well, they have done it, but at a really high cost.”  So that could be what we are facing here. 

But in the fleet of policies that are starting to emerge now, they are -- and in your paper, you -- again, I invite you to talk and expand your views on RPS and other kinds of things.  But there is at least an element of economic logic towards the use of a low carbon fuel standard versus a direct specification of picking a technological winner.  And similarly with the way that they are designing the efficiency programs are leaving -- they are creating incentives and sending the companies out into the field to pursue energy sufficiency programs without telling them exactly what has to be accomplished and where.

So, I think that is a little bit refreshing, but maybe you can tell us more about the ways that is sure to fill because for sure it will raise cost versus like cap-and-trade program that was just universal.  And I think the reality is that ultimately as economists, they take away our business card if we do not agree with the fact that prices will ultimately get the job done, but those prices can take a very long time and they may have to reach very high levels.  And, ultimately, they will get this done and maybe those high levels in that long time is politically unsustainable, so maybe that is why in a political economy framework, we need to couple price effect with some surgical policies to try to identify market failures.

In response to what Lee said, I just want to -- well, in fact, I think -- let me just draw your mind to thinking about the work by Morgenstern and Nelson and Harrington at RFF that looked at across a fleet of programs and found that, in fact, the cost of achieving regulatory programs that have used incentive-based approaches have not been, on average, more expensive than the average, they have been less expensive than were originally anticipated.  But there is more of a story there than that.  What typically happens is the total cost have been less than anticipated, the marginal cost have been less than anticipated, and the emission reductions have been less than anticipated in order to achieve those goals because something is going on in the baseline.  There is indigenous technological change or evolution of preferences or something is happening that is causing our economy to tend to become more efficient over time. 

And so I do not think that you can make a categorical statement that these programs always turn out to be more expensive than they think.  We think they are going to be.  They may turn out to be less expensive, but then the challenge is have we captured adequately that engine of change in the way we do our modeling work?

And finally, I just want to close with -- although I sort of -- in my original comments I said that the main -- with the way value I get from the EPRI study that David has done here, is that it allows us to do, look at a comparison of instruments, but I tend to put less faith in the number -- of absolute cost numbers that are identified rather than just seeing qualitatively why it helps make the persuasive argument of why you want to move to incentive-based approaches to achieve these social goals wherever and whenever possible, but not universally is my opinion.

But the other side, Lee brings up a very potent intellectual counterbalance to that is that economists cannot just give up on the notion of how much is it going to cost to achieve these goals because adaptation looms large.  And it is irresponsible on every sense of the word to not be paying attention to the opportunity for adaptation and to be doing a research and development on adaptation as a [indiscernible] because climate change now is inevitable and it is upon us.  At least, that is the way I interpret the science.  So for that reason then, to try to achieve a balance between adaptation research and adaptation measures and mitigation research and mitigation measures, we need to be informed by the kind of modeling work that David is putting together.

Samuel Thernstrom:  David, Dallas has invited you a couple of times to speak about RPS.  Do you want to accept that invitation briefly or respond in any other way to Dallas’ --?

W. David Montgomery:  Let me just say what came into mind in response to Dallas, and that can sit to the next round of whether I have captured what you were thinking of.

The idea of Renewable Portfolio Standard is to create a set of more or less market-based instruments; you actually create something you can trade.  But it only includes what are classified as “renewables,” and renewables generally mean wind, solar, and biomass.  I'm in a little trouble of being sure about wind, biomass as renewable.  Pennsylvania includes waste coal as a renewable resource so there is a certain amount of the politically correct resource, but coal with carbon capture and sequestration, which effectively removes 90 percent of emissions.  Nuclear power, which has no emissions, are almost never included as part of the Renewable Portfolio Standard. 

So it is a little hard to see what the policy problem is that the Renewable Portfolio Standard is trying to address other than creating a market for people who produce wind, solar, and a couple of other kinds of energy.  The difficulty is that, when that Renewable Portfolio Standard is binding and forces, for example, a lot of wind in the market and there is also an emission cap, the Renewable Portfolio Standard drives out in our modeling coal with carbon capture and sequestration.  So something that costs 50 percent more is forced into the market and replaces what would otherwise have been chosen under the motivation of the emission cap, which is a much cheaper way of getting to exactly the same result for greenhouse gas emissions.  And I would be more broad about it; I would say, “We have sulfur regulations, we have mercury regulations, we have NOx regulations.”  And all of those set up the incentive to choose the cost minimizing fuel and the RPS as kind of looking for a problem to solve, but forcing a particular way of meeting all of our environmental aspirations.

Samuel Thernstrom:  That is very interesting.  I do -- it is 3:25.  I do want to open up the floor for all of your questions; I'm sure you have many of them.  If I could, I just wanted to actually put one quick question of my own to Dallas before I do that.

This question of the “uncertainty” involved in the California system and the long time frame involved in the goals that are being set up and you did speak a few minutes ago about your confidence in the experience and skills of the California agencies in dealing with these sorts of problems in the past.  But I wonder if you have any more thoughts specifically on the problems that having such uncertainty in the investment climate pose.  I thought David’s remarks on that were particularly interesting.

Dallas Burtraw:  I really agree with David’s remarks on this.  I believe that uncertainty about the process undermines the incentive to make investments and innovation and the diffusion of that innovation into the field.  And so I think that the -- with the expression of the political will the agencies would do their job -- and if that was the end of the story, business could count on that as certain may be very expensive but it might cost a lot of investments to come to the fore.  But the problem is that those costs run up -- it is a preface for a potential undoing of the political will to keep going further.  So that is why it is really important to do these things cost effectively.  And Lee was talking about the engine of advocacy in pushing for a moral cause, but I tell you one thing that is really different about the climate debate is many of these advocates recognize that the potential enormity of what they want to accomplish on a global scale, they have embraced the idea of efficiency. 

So many of the national environmental organizations - and this is less true at the state level but increasingly true - really believe in cap-and-trade and believe in whatever mechanisms are going to get this done in an efficient way.  So there is an unusual alliance that is emerging in this context to think about the prospects of achieving these goals in an efficient manner.

Back to the point about uncertainty though, an appointed -- Dave only mentions that I did not get the chance to talk about was the role of a safety valve.  The California [indiscernible] community is not recommending the use of a safety valve but that was -- it is an instrument that was supported by some of us as an important feature that should be looked at in the future.

One problem with the safety valve is that in throughout almost all of Europe and through most of the state capitals, and so virtually everywhere outside Washington, DC -- and here it is true to 40 percent of the case -- it has been a totally politicized and polluted concept at this point because people hear the words “safety valve” and they are not thinking safety valve.  They are thinking a very low price and an escape hatch on compliance responsibilities for business.  That is all they see, and they see “red” when they think that. 

So sometimes people [indiscernible] than myself need to change the terms of the discussion so that we are thinking about cost management mechanisms in a way that are more robust than what safety valve gives us, I think, because it is just so firmly rejected in what to me, as an abstract economist, seems like in a rational way because it is just a tool in your toolbox and it is not predestined to be set at very low levels.  But we have offered a dynamic, symmetric cost measure mechanism: There should be a floor as well as a ceiling on emission allowances, the floor providing a certainty -- I mean, if you look at the experience of cap-and-trade programs in the world, the largest costs have been associated with unexpected price falls not on unexpected price increases. 

No, I'm not suggesting that is going to happen with CO2; I realize this is a different case.  But if you look at, for example, the SO2 program, Congress got nowhere near what they thought they were buying in terms of what they were willing to spend to achieve emission reductions for SO2 when the cost fell dramatically, but we were locked -- our feet were in cement with respect to the cap that was established by statute, and it took 15 years to sort of revisit the level of that statute so there was no sort of dynamic adjustment on the quantity mechanism.  What is historic about cap-and-trade, and really it has never existed previously, is that we have instantaneous feedback information in a summary statistic in one number in the marginal cost of control throughout society to achieve an environmental goal.  Never have regulators had that information before available to them in an instantaneous way, and you get that through the price index value of an emission allowance or perhaps the willingness to pay it in an auction.

And so, regulators have instantaneous feedback of what it is actually costing society to achieve a goal that they have set forward.  You do not have that type of instantaneous feedback on marginal benefits, but you do have it at a marginal cost.  And I think we need to think about cost management mechanisms that feed that back into the process to help us achieve our goals in the long run.

Male Voice:  Feed that back [audio glitch] levels [audio glitch] perhaps?

Dallas Burtraw:  Yeah.  The problem with the one-sided safety valve - and my colleagues at RFF who I love dearly have been developing this idea and championing it for years - but it is a one-sided mechanism, and environmental advocates perceive it as undermining the environmental integrity and undermining the incentive for investment and innovation.  And they are right.  It was we have thrown this into a model, that is the castic [sounds like] modeling, and shows -- I mean, it just falls right out of your intuition.  It develops and forms it very quickly that probabilistically you are going to hit that cap, and so probabilistically there is going to be more emissions than you thought there are going to be.  And what you are doing is truncating that one portion of distribution of payoffs for an investor in renewables who says, “Oh, well, there is a 10 percent chance that very high payoffs are not to you -- are not going to happen.”  And so it shifts the mean to expect the rewards from it during an R&D or innovation, but you can recover those expected values by putting on a price floor. 

And the interesting thing about this is that to the extent that an auction plays any important role in the distribution of emission allowances like fundamental feature of good auction design is a reservation price.

For example, most of the auctions on eBay will have this, and some of the hard lessons that were learned when there have not been reservation prices and the famous stories about the sales on banks in Spain in such a -- where there has been severe embarrassment of buyers and sellers because of the failure to have a reservation price.  So you have this natural mechanism for price floor that could couple very well to a safety valve and perhaps help us recover this debate of how to do cost management. 

Samuel Thernstrom:  Any [audio glitch] else, David or Lee, on safety valves and on price floors, or shall we open up?

W. David Montgomery:  I have always loved the idea of the setting the price floor and the price ceiling identical.

Lee Lane:  And we could even use a different word for it than a safety valve; we might call it a “carbon tax.”

Samuel Thernstrom:  Okay.  Fair enough.  I'm sure people out there have some questions at this point, and I think Evan is here with the microphone.  So if you have a question please raise your hand to be recognized and wait for the microphone.  We got our first question right here.  Please identify yourself when you ask your question.  Thanks.

Maya:  Hi.  My name is Maya from New York, interning with the Economic Strategy Institute.  I have a couple of questions directed to both Mr. Montgomery and Mr. Burtraw.

First of all, to Mr. Burtraw, bringing in to the discussion, I guess, the Civil War, I do not think that “morality” is a bad word in this discussion, especially in the climate change context.  So public policy in this issue, I think, is critical.  As far as the safety valve goes and the discussion about that, in all probability, is safety valve will have to be introduced, if anything, to appease a lot of policy makers and making the system more predictable and more reliable.  And in that case, what do you think can be done to prevent the inflation, I guess, of the safety valve permit prices, essentially saturating the market and driving down the cost of permits so that essentially the market becomes ineffective?   

And perhaps if putting a limit on the life expectancy of the permits introduced within this period could help achieve that, as in they would be only temporary for only a certain year, and then afterwards would be essentially invaluable -- taken out of the market if that gets some potentially help in that aspect?

And to Mr. Montgomery, I wanted to ask if in your calculation in the matrix about the GSP expectancy in the laws of welfare, do you take into account any potential, even modest potential renewable energy income, jobs added to the Californian market or any other sort of increase in productivity or any sort of financial gains that could be achieved through this policy?  And if that is calculated within the matrix and within the econometrics [indiscernible] very much.

Dallas Burtraw:  Well, I appreciate your view about the use of the term “morality.”  I would hesitate to use it myself, but I'm just telling you that in the state capitals and the public hearings we have had we are hearing that word. 

With respect to the idea of the safety valve and the limiting price fluctuations, limiting time on the time length or lifetime of allowances does not seem like the step on the right direction because I'm a huge advocate of banking.  I think that the accumulation of a large bank at the front end provides -- it was one way of providing certainty for firms and it also creates the political economy of this or to a program was that it created a vested interest than the longevity of that program because firms woke up to the fact that they were sitting on top of a valuable resource in the bank.  And if you talk to people at EPA, they were really surprised at the lack of litigation around Title 4 when they were first writing the rules to bring it to bare end.  They think one of the things that contribute to that was these firms realized, “Hey, this is not so bad.  We got US$100 million of allowances here.” 

And in the long run, in the context of a cap-and-trade program, that is going to be very, very expensive endowing firms with a bank at the front end as very little environmental consequences and can really contribute to the longevity of the program.  So I'm a big fan of banking, and so other things being equal, I would want to try to see that banking preserved for the long run so the investments and the return from doing innovation has a long-term reward.  But there is something that can contribute to trying to keep prices down in addition to the safety valve, and the regional greenhouse gas initiative folks and the ten northeast states have a provision that requires a minimum of 25 percent of the emission allowances or 25 percent of the allowance value has to be invested towards strategic energy purposes.  This is essentially mandating auction floors of at least 25 percent of the emission allowances. 

So this gets back to, well, is there a regulatory failure?  Is this raising the social cost of the program?  It may be raising the social cost of the program, but it is dampening the value of the allowance price to force these investments and use efficiency and renewable technologies in the northeast states.

So, for example, what we have done or what happened when Maryland chose -- became the tenth state joining the northeast states and we put in this -- and Dodge’s [phonetic] modeling of a fraction of the allowance value and use the efficiency -- we use the AC Triple-E estimates for how much efficiency is out there and we thought, “Well, the AC Triple-E has got to be pretty -- perhaps seeing a lot of opportunity which is there that we might all or not agree on were [indiscernible] are ready.  So we thought we would put on a pretty hefty penalty on that of 2.4 cents per kilowatt hours program cost on top of that.  And still, when we did this, we saw electricity prices fall when Maryland joined ReGGIe within the context of our model.

So now I'm starting to think, “Well, maybe 2.4 cents was not enough; maybe we should have done it 4.8 cents.”  But the point is that, in any case, there is an important feedback if there is an investment in end-use sufficiency in other goals with some portion of the allowance value, and there might be a variety of reasons to do that; we heard a lot. 

In California, there is a blocking coalition from the Environmental Justice Community, and so one of the ways to appease that community was to just simply recognize that for 10 or 20 percent of the state that is low-income, they get that in both hands.  When it comes to adaptation measures, slow-income people that are getting creamed by the effects of climate change; when it comes to cost of mitigation like the increase in cost of energy, the slow-income people are getting creamed there also.  So one of the design characteristics that are recommended is that some portion of allowance value should be reinvested to help achieve energy efficiency through end use.  And those types of investments have a cost, for sure, and they may not even be the least cost way to achieve carbon goals from the social perspective.  But you are not just buying carbon reductions there; you are buying a social contract in terms of softening the blow to low-income households.  And that will keep allowance prices lower -- I mean, thereby construction within a model to achieve more energy efficiency is going to help lower the allowance cost.

W. David Montgomery:  Yeah.  On a completely different question, yes, in our calculation of impacts on gross state product down on broader measures of economic welfare in California, we include all of the effects on the economy, both the increases and investment and output for renewable energy or equipment that goes in the houses or businesses to improve their energy efficiency, as well as the effects of, say, reduce the output of the chemicals of the oil industry.

The models, basically, ask the question: how are the scarce resources in the California economy utilized when this policy is imposed?  And so we see a certain amount of shifting within the economy, but the net effect is actually a reduction in productivity because of the higher cost of energy pervasively in the economy for producing the things that people actually want, which is not energy but the goods that you get from businesses that use energy to supply things.

Samuel Thernstrom:  Other questions?  Right here, sir, on the aisle.  Evan?  No.  Right here.  Yeah.

David Moore:  Hi.  My name is David Moore with Tudor Investment.  First, I have a question for Dallas.  I wanted to see if he could talk a little bit more about the details with the offsets.  When you say the offsets coming from in the California -- [speaks away from microphone] -- California-based offsets, domestic?  [Speaks away from microphone] international offsets [speaks away from microphone].  And if I could just throw another simple second question to David, is a CRA report accessible [speaks away from microphone]?

Dallas Burtraw:  Well, actually, I'm not going to talk about offsets a great length.  Now, apparently, it is not really my area of expertise except to say that California, really, that we are recommending unlimited offsets, but that would be staged in a sense that they be held to a very high standard and that that standard be expanded.  And it is going to require some time in order for that to be expanded so various types of certification mechanisms could be put in place.

But I'm not going to go into it more because I know that David’s report talks about offsets, and so I want to invite you to say what -- to address this issue.

W. David Montgomery:  Thanks.  Yeah, we only did a limited amount of modeling of the offsets in this report because of, well, first of all, the open question that Dallas is describing, which is what kind of rules California would set for qualifying the offsets, what type of offset, what type of projects they would include as being eligible for offsets, and then the huge question of whether California will allow projects outside the state to provide offsets against CO2 emissions in the state.

The one we did was forestry, and we took the assumptions from the Climate Action Team for the amount of offsets that could be generated through forestry as a kind of essentially free reduction in emissions to try to look at whether that is how much difference that set of offsets would make in the overall economic costs.  And they made a very substantial difference; but even having that at zero cost was not enough to bring the costs for California in total down.

One other observation about offsets that I think also, and actually occurred to me in listening to the analysis: First set of comments was, if we put together the three parts of the picture, which are the existence of a cap-and-trade program which puts as a price on carbon in there for it provides the demand for offsets -- I mean, that will motivate the projects kind of across the board that are qualified.  But then at the same time, we have a set of regulatory measures which mandate reductions in emissions.  They are actually going to drive down the price of carbon and essentially are going to serve to shrink the incentive for the offsets.  So the kind of the regulatory measures in some ways may be competing with creating a big enough market for offsets because the market for offsets is going to be what the cap-and-trade system is freely looking for to the extent that the cap is satisfied through the results of regulatory measures.  There is going to be less demand for the offsets. 

We will probably continue this year in looking at that in a lot of areas because EPA and others have developed marginal abatement cost curves for lots of different kinds of things in trying to apply those.  If we can get some regional detail, the California situation would tell you a lot but we just tried one example on offsets in what we did.  And on the report itself, it is available at the EPRI website if you just go to www.epri.org -- com.  That is right; epri.org is somebody different.  www.epri.com.  I believe they still have on the lower right hand side of their homepage a link to the California Climate Study, or you could go to the CRA website and actually if we can light up the computer again, I’ll put our URL up there that people can see and write down.

Samuel Thernstrom:  Other questions?  In the back there.

Janet Peace:  Hi.  I'm Janet Peace.  I'm an economist with the Pew Center on Global Climate Change.  I just have a couple of questions for David on your model, if you do not mind.

First of all, thank you for a very nice presentation to all three.  But back to my questions, David, in terms of your model, I'm wondering if you could talk about two different things: First, the advanced technology that we know we need to achieve the significant reductions that are being put forward, what kind of technology did you include in your model?  You mentioned biofuels.  I'm wondering whether you included things like plug-in hybrids, algomats [sounds like], stock separation technology that is coming to the forefront.  What kind of technologies did you actually include?  I know that is kind of difficult, but if you could talk about that for a minute. 

And second of all, in terms of the impact on contract shuffling and leakage, I'm wondering if you included or if you did any runs that had the six dates and two provinces, that particular regional effort that is coming into the play, or even the fact that Oregon, Washington, BC, Manitoba, et cetera are looking at targets and timelines, whether you were able to get back to those type of constraints?

W. David Montgomery:  Thanks.  In terms of the advanced technologies, as we looked out -- we include on the electric power sector just about everything that you could imagine as a generation source.  The biggest issue there is nuclear and California’s unwillingness to see nuclear in the mix, and so that was something we assumed in our fair trade case.  When we took off the limit on nuclear, it reduced the cost significantly of achieving the 80 percent targets in -- it significantly reduced the cost of meeting the 80 percent targets.  We did not do a lot more experiments with bringing in new technologies because that kind of made the point within the limited number of sensitivities we were trying to do, which is that opening up to all of the available technologies and finding lower cost sources of generation with zero emissions can make a huge difference to the costs.

Plug-in hybrids is something we tried to represent in a solution for the transportation sector because that is clearly important, and it was a technology that we brought in which I tend to think it was biofuels but actually what it was, was it was represented as a technology that provides a zero carbon fuel that has learning by doing, and so that cumulatively the more you invest, the faster it can be deployed.  That made a very big difference.  Taking that out and putting it in had a huge impact on what the costs are likely to be.  So again, that clearly shows that that is very important.

We also built a tremendous amount of energy efficiency improvement into our baseline so we have essentially electricity consumption growing at about the rate of population growth in California.  Something there is a kind of folklore in California that that has been what California has accomplished through its policies in the past so we assume that it continues in the future.  I think that it is insanely optimistic, but it does wrap into the baseline a whole lot of the things that you are talking about that will take new technology in order to be able to do.

As far as the impact on leakage, yes, I think that is about the only solution, which is a source-based as in probably the way that Dallas was describing it: not a load-serving entity base but a source-based cap-and-trade system for emissions from electric power generation covering the entire Western Interconnect; that is about the only way you can do it.  It is just one -- there is not much connection across the Rockies so you could do it in the West; you would not have to do it the whole country.  I think anything -- but I think that is what it takes to do it. 

We did not model that explicitly because we did not actually do -- again, we were limited in what we could accomplish in the time and the resources, and we decided to complexities of trying to figure out how California would fit into a broader federal system or something we wanted to take up this year rather than last year.  So we did everything California alone.

This one, in particular, it is pretty clear that that is a way of dealing with the leakage problem.  There is going to be a lot of other interesting implications if you started thinking about how to do that right for how do the different regions set the caps, who is going to be trading with whom, which direction are the benefits going to be flowing, is it actually -- I mean, sure, we can get rid of leakage but depending on where the other states set their caps, California may be spending a lot of money to buy permits from those other states rather than doing emission reduction. 

In fact, we did do a little bit of that on ReGGIe.  We compared our carbon prices which were about $100, say, $40 a ton.  By 2015, in some of our ca