What’s the Value of an Energy Economist? Peter Davies. Special Economic Advisor, BP plc Address to the Annual General Meeting of the British Institute of Energy Economics London, 12 November, 2007
What’s the Value of an Energy Economist? Peter Davies BIEE AGM, November 2007
It is a great pleasure to have the honour to address the AGM of the BIEE today. It must always be difficult to chose a subject for such a talk. As I was thinking about this I was drawn back to a conversation I had some years ago with some non-economists from within the industry. During our discussions they asked: “Well what is the value of an energy economist?” I thought then and I still think now that this is a very good question. I feel that this is an occasion when it is acceptable, if not required, to be somewhat reflective about energy economics. This is particularly apposite for me. As some of you will know, I stepped down as Chief Economist of BP in September after 17 years in the role and will be retiring from the company next year. At such a time one tends to reflect upon the impact that energy economics can have and has had. Some have said that energy economists are well known for being wrong for the right reason. I am not certain whether this is a complement. I suppose that it at least implies that economists apply logical analysis to energy problems. It may also be the best that one can realistically expect in an area where the future has proved unpredictable and where nobody has proved to be right all of the time. As a result of this I thought that it would be valuable tonight to address the question “what is the value of an energy economist?” I spoke briefly on this topic at the IAEE annual conference in New Zealand early this year. I would like to pick the issue up again tonight and to develop it somewhat further. Broadly I believe that energy economists only have value if they can improve decision making – whether in terms of setting policy or in making investment decisions. So what do we as energy economists actually know?
1
I could approach this in typical economist’s fashion. When asked any question the economist’s first answer is usually – “it depends”. It always does. And when this fails to convince the next answer is often “it is too soon to tell”. It usually is. If I did, this would then be a very short presentation. To be more constructive, I want to approach this by addressing 5 of the key issues or themes in energy economics today. I want to try and identify what energy economists know – and especially to identify where and why this may differ from the conventional wisdom or other views held elsewhere. Issues • Resources • Demand • Climate change • Energy Security • Oil prices
2
The five issues I want to focus on are: 1. Oil resources. I particularly wish to address rigorously the issue of so-called peak oil that we continue to hear about in the media and from some particularly vocal geoscientists 2. Second, energy demand. What do we know about what drives it and can we expect to see any significant changes of trend in the near future? 3. Third, climate change. What do energy economists add to the analysis and debate? 4. Fourth there is energy security. It tends to be an issue related to perceptions and fears. What can the economist bring to this discussion? 5. And finally oil prices. Was the recent price rise predictable? If we had known the supply and demand outcomes with any degree of accuracy, could we have predicted the price? What can we say about future prices?
2
The Hubbert Curve Fallacy
You Are Here
Oil production depends upon: • Resources in place • Investment but also • Price • Technology • Taxes • Government policies • Demand – at a global level
3
Let’s start with oil resources. I am certain that we as energy economists know that peak oil analysis that uses Hubbert curves to predict the imminent peaking of world oil production is wrong. Such analysis is conceptually flawed. Most proponents of peak oil believe that the amount of ultimately recoverable oil globally is fixed and knowable and that world oil production peaks when 50% of this oil has been produced. This is based upon the central limit theorem that concludes that a set of independent random variables – oil fields – tend to generate a Gaussian distribution or, to us economists, a Normal distribution. In other words oil production should follow a bell curve. But this is flawed. The amount of ultimately recoverable oil is not fixed, nor is it knowable. It is not simply a function of geology and physics, but also of other variables – often economic variables – including price, demand, technology, taxes and other government policies. It is not a random distribution. A Hubbert curve, a bell curve, is not inevitable and, at a global level, not even likely. A skewed distribution is more probable. Coincidentally, the Hubbert analysis did broadly apply to US oil production in the middle of the last century: it was at the level of the US not the world, so demand did not matter; prices were low and stable; there was little technological development onshore in the US as the Middle East was surging ahead; and taxes and policies were also relatively constant. In such circumstances the central limit theorem can reasonably be expected to apply or to be an approximation – and it did. But that is not the case at a world level and is not the case today even in the US where production is now higher – and with a longer and stronger tail - than Hubbert’s analysis would predict. Technology and prices have kicked in. The more radical peak oil analysts believe that they know how much oil is ultimately recoverable. They usually assume a low figure and almost coincidentally the world happens to have produced close to 50% of this amount already. Thus the so called inevitable peak is going to be upon us very soon. The precision of their estimates is uncanny. One, Professor Ken Deffeyes of Princeton, was confident enough to predict that world oil production would peak on Thanksgiving Day - 2005.
3
“The Campbell Fallacy” X
Today
1988 Peak Oil Forecast
36 million b/d
4
Peak oil devotees have been predicting the peak for some time. One, Colin Campbell, had predicted that it would occur in the early 1990s and that world oil production would by now be in precipitate decline. In fact he had predicted that world oil production would now have declined to 36 million b/d, well under half of current levels. Economists who have been more confident about resources and supplies have unequivocally won the empirical debate so far. But this is not a black and white issue even though some like to portray it that way. The more avid proponents of peak oil have polarised the debate. Those who disagree that the peak is imminent are sometimes portrayed as ‘flat earth economists’ who believe that oil supplies will be plentiful for ever without constraint. I hope that most economists are more balanced. They realise that: 1. The amount of ultimately recoverable oil is not fixed. 2. Oil resources deplete and production has already peaked in a number of provinces or areas including the US and the North Sea – and the OECD as a whole. The world’s oil resource is maturing. 3. World oil production will peak at some time. Nature produces only a minimal fraction of the oil we produce and consume each year. And the more narrowly oil is defined, the earlier that peak will be. 4. There is a distinction between ‘conventional oil’ and liquid hydrocarbons. Consumers consume liquid hydrocarbon products rather than conventional oil. They don’t care if it is from a conventional oil field, a GTL plant or oil sands – just as they don’t distinguish between North Sea gasoline and, say, Angolan gasoline even if they could identify it. The broader oil is defined, the lower is the likelihood of an early resource constrained peak and the lower is the probability of a symmetrical bell curve global oil production profile. World oil production will in time peak for one, or more, of several reasons:1. World oil supplies could be seriously disrupted such that it becomes physically impossible to produce or supply as much oil as in the past. This is conceptually possible but economists have
4
little to add to the prediction of such an extreme outcome. This is essentially a political issue. 2. World oil supplies could peak due to a lack of investment in oil production capacity, especially in the areas where resources are most abundant. This could occur even if oil resources are geologically proved and economically viable. Access to oil resources for commercial development has become more restrictive and fiscal terms have become more severe in recent years. Some National Oil Companies are either unwilling or unable to expand capacity, despite adequate resources. Globally this is a risk over a 10 year horizon or longer. 3. World oil production could peak because oil demand peaks. I will discuss demand more in a minute. However, most scenarios that model long term sustainable climate outcomes imply a peaking of world oil consumption within a 20-30 year time frame. This is essentially an environmental policy judgement rather than economic analysis per se. 4. And world oil production could peak through resource scarcity or exhaustion. There can always be concerns about serious disruptions. The critical issue today is probably one of the level of investment. The slightly longer term issue is as much one of the severity and form of climate change policies as of resources. I am personally cautious about investment levels over the coming decade. I also think that oil demand could slow and start to peak within a generation. On the resource issue, as economists, we can dismiss the inevitability of bell shaped curves and the ‘50% rule’. It is then a matter of the empirical evidence – taking into account the impact of technology and prices on future supplies. The empirical evidence still points to a global adequacy of oil resources and reserves for a generation. GDP and Energy Consumption Index 160 1990=100 150
The link remains intact …….
World
140
GDP
……. but varies regionally
130 120 110
Energy
100 90 1990
150 140
OECD GDP
130
1993
FSU
100
80
1999
2002
China
400
GDP
200 180
GDP
300
Other Non-OECD GDP
160
250 Energy
140
Energy
60
150
90
50
100
100
19 90 19 93 19 96 19 99 20 02 20 05
200
100
19 90 19 93 19 96 19 99 20 02 20 05
70
19 90 19 93 19 96 19 99 20 02 20 05
Energy
2005
19 90 19 93 19 96 19 99 20 02 20 05
Energy
1996
350
90
120 110
110
120
5
The second theme is energy demand. We know as energy economists that, other things being equal, world population and economic growth will ensure that energy demand – and consumption of oil, gas and electricity - will continue to rise, albeit slower in aggregate than global GDP. The linkage between demand growth and GDP growth varies around the world, as a result of different stages of economic development, the composition of economies and resource endowments. But the linkages are still strong. Income elasticities of energy demand tend to be higher in developing economies than in the more mature post-industrial economies of the OECD.
5
Oil Consumption ‘S Curves’
35
US
China
10
15
S Korea
UK
Japan
Oil consumption, barrels per person
30 25 20 15 10 5 0 0
5
20
25
30
35
40
Real GDP, thousand US$ per person 6
It is valid to draw ‘S curves’ – plotting energy or oil consumption per head against GDP per head over time. The developing world is moving up the steep part of the curve. Transition economies are moving back towards the curve. OECD economies tend to be on relatively flat sections. So what does this tell us? It says that, other things being equal, continued economic growth will continue to drive energy consumption growth, especially in the developing world – including China and India. In fact, as the world economy accelerated in the early years of this century, the faster economic growth was concentrated in non-OECD countries and was especially energy intensive. The energy growth momentum has remained strong GDP and Primary Energy Growth Growth
Primary energy consumption
GDP (PPP)
6% 5% 4% 3% 2% 1% 0% 2002
2003
2004
2005
2006
7
Meanwhile economics should also tell us that, as energy prices rise, energy consumption growth should slow. And the evidence is that there has been a price effect. In both 2005 and 2006 global energy consumption growth slowed relative to GDP growth. Also, where energy consumers have been shielded from world price rises, energy consumption has been the most resilient We can also observe that the price effect is larger and more rapid when there are substitutes – choices. Fuel switching occurs more when energy users have a physical and economic choice. This was the case when residual fuel oil was substituted in the early 1980s and more recently, for example, in the US natural gas market. High US natural gas prices led to switching at the margin from gas into coal and oil – and more recently out of oil again as gas prices have fallen back relatively. And gas intensive industrial production has been weak; there has been investment in other countries where gas and other costs are lower.
6
But price effects on oil consumption should be expected to be smaller than in the 1980s. The potential for substitution of non-transport oil use is less. There is also some evidence that in the US, the price elasticity of demand for gasoline has declined. Regardless, gasoline consumption growth has slowed in the US as prices have risen. Nevertheless, in richer countries, gasoline is no longer a price sensitive luxury and has instead become more of an ‘essential’ commodity that is not willingly foregone. Overall, this tells us that energy demand growth still has underlying momentum, especially in the developing world. High prices will slow the trend rate of growth but, so far, the evidence is that this will be insufficient to create rapid adjustment and lower prices in the near term in most energy markets. Climate Change Carbon Emissions • Science widely accepted despite many uncertainties ….. • ….. but emissions show no sign of slowing
Index: 1990 = 100
140 World
130 120 110
Annex 1
100 90
Annex 1 Ratified
80 1990
1994
1998
2002
2006
8
So, turning to issue 3, what can we say about climate change? First we have to listen to the scientists about the science. The consensus is now extremely strong and well founded that, despite many uncertainties, human activity is increasing carbon concentrations in the atmosphere and that the world is as a result getting warmer and will warm further. This is not inconsistent with the reality that there are still many uncertainties about both the science and the empirical assessments. The conclusions I have already made on energy demand imply that carbon emissions and concentrations have risen and will continue to rise for the foreseeable future, other things being equal, until mechanisms are in place that effectively reveal a cost of carbon and succeed in incentivising technological advances and investments in low carbon sources of supply and more carbon efficient consumption. There is a long term bust unless something happens. There is a fundamental inconsistency between current energy demand trends and almost all estimates of greenhouse gas emissions consistent with a sustainable climate.
7
Key Climate Questions • Mitigation/Abatement versus Adaptation - Some adaptation inevitable - Long term cost benefit calculations sensitive to assumptions - Precautionary actions • What instruments? - Cap and trade versus carbon taxes - Role of regulation
9
There is a valid debate about the economics of mitigation or abatement today versus adaptation tomorrow – and what balance there should be between them. Our extrapolations of energy demand and greenhouse gas emissions and the knowledge of scientists make it certain that the world will warm further whatever happens and probable that there will be other more serious disturbances to the climate. Some adaptation will be inevitable. But do we know the costs and benefits of abatement? There have been some very serious attempts to make estimates – aiming to calculate the discounted differences between two very big numbers some 50 years into the future. The IPCC and the Stern Review conclude that the costs of action are lower than the costs of waiting. Unfortunately small differences in assumptions make big differences to estimates about net benefits. We also have to consider distributional impacts. It is very difficult, if not impossible, to be certain or precise. The overall answer may be that the net cost of mitigation or abatement will be small. I am a believer that technology will respond to incentives. Hopefully these incentives will be in the form of a global price for carbon, but in reality they will inevitably also include command and control mechanisms. We cannot be absolutely sure that the cost of abatement will be low or minimised and we should accept that there is a wide range of uncertainty. Meanwhile, there is a case for taking precautionary action today, that is measures that are proportional to the risks that we face. There is the option to bear costs today, just in case and especially in case there is a catastrophic outcome. Or alternatively we could wait in the hope that the costs will be lower later. Technology can be reasonably expected to progress in face of incentives and to reduce future costs of abatement. But there is also a cost to waiting, not just benefits foregone. Higher carbon emissions will remain in the atmosphere for hundreds of years. There is a risk of a very high cost catastrophic outcome. There is a thus a cost of waiting, which could be very high, that needs to be set against the lagged benefits of future technological advances. These concepts should be very familiar to all option theorists. We cannot make a cost-benefit analysis with a high degree of certainty, but I think that there is a good case to take precautionary action today to reduce the possibility of very high costs. There are non-negligible risks of very high cost outcomes. This conclusion applies even if many will try to free ride. And so what should we do about climate change? I hope that as economists we would all agree that we need to reveal a market based cost of carbon globally that will eventually deliver emissions
8
and concentrations consistent with a sustainable climate. Market based responses are most likely to deliver lowest cost solutions. Command and control interventions are only justifiable when there are principal-agent problems (as with building standards), when transaction costs are prohibitively high or where there are other externalities. How should the cost of carbon be revealed? The choices are cap and trade schemes or a global carbon tax. In economic terms they should be equivalent, if established soundly. A cap and trade scheme delivers a certain outcome at an uncertain price. The tax delivers an uncertain outcome at a given price. As we are primarily concerned with achieving a volume outcome – i.e. a sustainable climate with given caps on carbon concentrations – cap and trade appears to me to be preferable. It also has the benefit of stimulating cross border emission trade to ensure that the reductions occur where the costs are lowest. The economic consensus a few years ago was to prefer carbon trading. SOx trading in the US has generally been considered to have been successful at achieving a desired environmental outcome at a low cost. The initial experiences of trial carbon trading schemes such as that within BP were also seen to have been successful. However, the initial experience of the European Emission Trading Scheme (ETS) raised some doubts. Over-allocation of permits and gaming resulted in carbon prices ending up at zero. Allowance allocation mechanisms were seen as unfair and generated significant windfall profits to some companies. Partly as a result, the economics fashion has swung back more towards carbon taxes. The balance of arguments depend upon transactions costs, equity and other distortions. Auctioning of permits and greater experience about emission levels will make the ETS more effective going forward. Carbon prices are now positive again. It remains unclear what the optimal global tax rate should be, and if it can be effected. Meanwhile current energy taxes are often ‘carbon blind’ and need to be restructured before a new carbon tax can be effective. I continue to prefer cap and trade, but the economic arguments are finely balanced. But I have other concerns. This is really a second best problem. We are very unlikely to achieve the first best solution of a global cap and trade system or a global carbon tax soon. At the same time there is a further complication that energy prices are not market determined in many countries. So what should we do today? Is the second best solution a move towards the first best? Or is there a fundamentally different approach? I do not think that we can be certain about this, but I find it difficult to favour any actions that are not consistent with the first best solution. It is imperative that the effective cost of carbon is not zero and that it is the same or very similar for all greenhouse gas emissions from all sources, regardless of sector or geography.
9
Energy Security • Normal disruption vs catastrophe
• The role of insurance - stockpiles
10
Fourthly, what can energy economists say about energy security? We know that disruptions and uncertainties have a cost. The world is an uncertain and in some ways an unstable place. The balance of the world’s energy geography is shifting and the proportion of energy consumption that is traded across international borders is increasing. It is natural that consumers are concerned about the security of energy supplies, just as energy producers are concerned about the security of energy demand and markets. For the last several years we have had energy security. Energy supplies have been maintained, despite political disruptions such as the demise of the Soviet Union and wars in the Middle East. Global energy supplies were also maintained despite the devastation caused by Hurricanes Katrina and Rita in 2005. The key was that markets were allowed to work. Markets have been able to ensure the availability of energy supplies in face of a very wide range of circumstances. Markets provide energy security. The historical tendency has been for energy supplies only to be unavailable when markets have not been allowed to work. We should be aware however that there are broadly two types of potential disruptions to energy supplies: first there is what we may call ‘normal disruptions’; and secondly there are ‘catastrophic disruptions’. Catastrophic disruptions are something bigger and more disruptive than recent hurricanes and even wars. While we know that the best form of energy security in face of normal disruptions is in fact markets, we cannot be absolutely certain that markets will always be able to maintain supplies of energy in face of massive or catastrophic disruptions. Rather, by definition, such catastrophes are so big that markets cannot handle them. In such circumstances non-market interventions will be required to allocate available energy supplies on an equitable basis. As economists we can point out that it is possible to buy insurance that can reduce these impacts or costs – for example by building strategic or emergency reserves or spare production or generation capacity. I would argue that the choice of the amount of such insurance is ultimately a public policy choice. We can also highlight that in the event that there is catastrophic disruption, additional interventions will be required, potentially including command and control measures. It is better if such measures are not knee jerk responses. It is better to think about them ahead of time, even if the precise form of catastrophe cannot be predicted. Consideration of unintended consequences can only be
10
valuable. As economists we can help quantify the costs and benefits of both buying insurance and of command and control interventions in face of catastrophes. This should help policymakers to make rational sound decisions. And finally, energy prices – and especially oil prices. Oil Prices Monthly Average Dated Brent
100
?
90 80 US$/bbl
70 60 50 40 30 20 10
19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07
0
11
The price of oil is now in excess of $90 per barrel. 10 years ago prices were heading downwards towards $10. I know of no oil forecaster, whether an economist or not, who predicted the path of oil prices over this last decade. Some forecasters in the early 1980s were predicting $100 oil, as, I suspect, were some peak oil theorists. At best they were right for the wrong reason. As economists we should admit that we cannot forecast oil prices with any degree of accuracy over any period, whether short or long. No oil price model has proved reliable over anything but short periods. There are simply too many variables and uncertainties. Many of the key parameters are unforecastable. But I would assert that we economists can potentially be less wrong and more insightful than other forecasters – or at the very least be wrong for the right reasons. We can justifiably ask why oil prices have risen so far and to what degree oil supply and demand forecasts were wrong.
Oil Market Forecasts Forecasts of 2000-2005 Growth
8 EIA 2000 Actual
Million b/d
7 6 5 4 3 2 1 0 Oil Consumption
Non-OPEC Production
OPEC Production 12
11
The supply/demand forecasting errors were surprising. A review of the IEA, EIA and other forecasters generates a consistent set of answers. If we look at forecasts made around 2000 for 2005 what went wrong? 1. First oil demand estimates were reasonably accurate, although the cycle was inevitably impossible to predict. 2. Non-OPEC production was actually under-predicted. Output in the mature areas such as the US and UK declined faster than expected, but this error was more than offset by the strength of Russian production growth. 3. The biggest errors related to OPEC capacity. The politically driven production and capacity declines in Iraq and Venezuela were unforeseen and almost certainly unforeseeable. Also, most forecasters over-predicted capacity growth in the rest of OPEC. In reality such investments were widely delayed. The rise in oil prices from around $30 to $60 from 2003-5 was generated by a lack of spare capacity and increased geopolitical concerns. More recent price rises have followed OPEC production cuts in late 2006 and early 2007 and renewed concerns about geopolitical threats to oil supplies. I find that the current fashion to describe the ongoing oil price surge as ‘demand driven’ to be puzzling. Oil consumption growth has remained at or below trend. The forecast errors on OPEC capacity have been mainly politically driven. But even if we had been able to predict with total accuracy the levels of consumption and production today, would we have been able to predict oil prices? I think that such information would have been insufficient – which is why most models fail. However, if we had perfect foresight of the spare capacity data, we would probably have been bullish on prices. But spare capacity has recently increased and prices have surged. The relationship between spare capacity and price is not a reliable indicator of oil prices. At the very least it depends on why spare capacity has increased or decreased. Recent OPEC production cuts raised spare capacity – and tightened the market and have been a causal driver of higher oil prices. So good forecasts would have helped, but I believe that they would have been insufficient. We would also have needed to understand OPEC’s policy framework for both production and investment and how it would evolve. We would also have needed to have known about geopolitical risk. Good energy economic forecasting was a necessary but not a sufficient condition for predicting today’s oil prices.
12
Oil Prices Oil supply curves
SRMC
p Depletion
High cost unconventionals LRMC
Taxes and royalties LRMC
Technology Exploration Success
Low cost OPEC oil SRMC
q 14
But what can economics add to our understanding of future oil prices. Economics tends to say that prices will eventually move towards long run marginal cost in competitive markets. However, the oil market contains a cartel-like producer group and tax authorities that apply royalties and set special and often variable tax rates on production and consumption of oil. I am not even sure that I can define or measure the marginal cost of oil production. Is it production costs in Saudi Arabia where the incremental barrel is likely to be produced? Or is it deepwater or heavy oil production in non-OPEC countries? Or is it the cost of biofuels in Brazil, in the US with or without agricultural subsides, or even elsewhere? In other words which supplies are truly marginal and which are intra-marginal? Oil is exceptional in that we can even discuss whether the lowest cost production – Saudi Arabia – is the marginal supply. And then how do you allow for technological advances, cost cycles, taxes and depletion? Even if we agree on the concepts, we then have measurement problems. How do we actually calculate costs in the current world of cost inflation and capacity constraints in the contracting and development segments. And where will the demand curve cut the supply curve and how will this shift? We can observe that, in the very long term, commodity prices do not tend to trend upwards in real terms. But there can still be very long periods – decades - of trend price rises. We can also observe that, generally, short to medium term supply curves are upward sloping, while long term supply curves tend to be relatively flat. For oil that essentially requires that in the long run technological progress needs to match depletion. It can be argued that this has been the case for the last 100-150 years. And so for oil prices to rise over the long term, the requirement is that depletion begins to overwhelm technological progress; that other costs such as taxes rise; or that OPEC policies or its effective market power change. In other words the curve could slope upwards or shift upwards, or OPEC may exert more market power. I think that we could reasonably argue that in the past, that is in the 1970s, OPEC’s market power and policies changed. I also believe that OPEC’s more recent behaviours contrast sharply with those in the 1990s. Where does this leave us? Some would simply conclude that prices rise when oil is scarce and fall when it is abundant. Others make observations such as the more prices rise and the further they rise, the more likely they are to fall eventually – and vice versa. This is all more obvious than insightful or analytical. I would claim that economics tends to offer caution on long run oil price prospects. At the present time it may be that depletion has accelerated, that taxes have risen across the board and that OPEC is both more willing and more able to sustain policies that deliver higher
13
prices than in the 1990s. But, as recent data has begun to indicate, economic forces will deliver some form of adjustment. Oil price caution is still economically rational in the long term, but the indications are that this could be an extended cycle and that a higher tax take will ensure that the long run marginal cost has risen and therefore that average oil prices will not return to and stay at the levels seen in the 1990s. These are my 5 points. I have tended to be oil-centric. I have not been able to address all the key energy issues. For example, I have not discussed the benefits of energy market liberalisation, electricity economics, renewable energy, the role of financial markets or the relationship between inventories and prices. I am sure that each of you would add another different issue. However, I believe that we can still agree that good energy economics can help provide sound insights and analysis and help to ensure that good decisions are made. We should remember what we economists don’t know, as much as reminding others what we do know. So I think that energy economists have a value. But I suspect that we can only measure the value by what others are willing to pay us. Energy Economics: Final Thoughts • Economic forces matter as much as physical and geological factors • Cost benefit analysis can help promote rational decision making • Energy markets work • Oil prices are impossible to predict accurately – but caution is appropriate
15
And so what should we conclude? I hope that what I have discussed confirms that economic analysis helps address many of the fundamental issues in the energy world today and in the future. Sound economic analysis should therefore be valuable. I would like to leave four final messages from energy economics as I see it: 1. Nothing is fixed and solely reliant on physical forces and factors. Economic forces also matter. This especially applies to energy resources. 2. Second, energy economists can help frame questions - such as climate change and energy security - in economic terms, especially using cost-benefit analysis, and can therefore help to improve decision making. 3. Energy markets work. Scarcity and abundance drive prices and prices impact consumption and supply and ensure availability on a continuous basis. Price effects may take a long time to work through in a sector that is capital intensive and has long lead times – but they do work through. As energy economists we should be saying this again and again. We should be
14
focusing on time lags 4. Oil and energy prices are impossible to predict accurately. However energy economics leaves most economists somewhat cautious about long run price prospects. Thank you very much.
15
