All Episodes

[Episode #32] – Resources and Economy


The notion of “decoupling” energy consumption from economic growth has become vogue in policy circles, but how much evidence is there that it’s really happening? If the energy intensity of our economy is falling, are we sure that it’s becoming more efficient, or might we just be offshoring energy-intensive industries to somewhere else…along with those emissions? If energy reaches a certain percentage of total spending, does it tip an economy into recession? Is there a necessary relationship between energy consumption and monetary policy? Is there a point at which the simple fact that we live on a finite planet must limit economic growth, or can economic growth continue well beyond our resource consumption? Can the declining EROI of fossil fuels tell us anything about the future of the economy? And can we have economic growth using clean, low-carbon fuels, or might transitioning to an economy that produces zero net new carbon emissions put the economy into recession and debt?

To help us answer these thorny questions, we turn to an expert researcher who has looked at the relationship between energy consumption and the economy over long periods of time and multiple economies, and found some startling results with implications for the Federal Reserve, for economic policymakers, and for all those who are involved in energy transition.

Guest: Dr. Carey King performs interdisciplinary research related to how energy systems interact within the economy and environment. He is a Research Scientist at The University of Texas at Austin, and Assistant Director at the Energy Institute. He also has appointments with the Center for International Energy and Environmental Policy within the Jackson School of Geosciences and the McCombs School of Business. He has both a B.S. with high honors and Ph.D. in Mechanical Engineering from the University of Texas at Austin. He has published technical articles in the academic journals Environmental Science and Technology, Environmental Research Letters, Nature Geoscience, Energy Policy, Sustainability, and Ecology and Society. He has also written commentary for American Scientist and Earth magazines discussing energy, water, food, and economic interactions. Dr. King has several patents as former Director for Scientific Research of Uni-Pixel Displays, Inc.

On Twitter: @CareyWKing

On the Web:  Carey King’s website

Recording date: November 28, 2016

Air date: December 14, 2016

Geek rating: 8

Chris Nelder: Our guest today is Dr. Carey King a research scientist at the University of Texas at Austin and assistant director at their energy institute. He has a Ph.D. in mechanical engineering from the University of Texas at Austin and he performs interdisciplinary research related to how energy systems interact with the economy and the environment. He has extensively published on the subject and it's a great pleasure to have him on the show. So let's bring him into the conversation now. Welcome, Carey, to the Energy Transition Show.

Carey King: Thank you. Thank you Chris.

Chris Nelder: So I'd like to just kick this off by delving straight into the data on energy intensity. Does the declining energy intensity of an economy tell us that resource scarcity is not in fact an important issue and that economic growth can continue without a corresponding growth in our consumption of resources.

Carey King: So energy intensity by itself I don't think tells you too much about resource scarcity, although, we could argue maybe it tells us something. One way I try to think about energy issues is to think about how various trends have changed over time. And one of the longest time-series about energy intensity that I'm aware of is about England and the United Kingdom over the last seven hundred or so years. And when you look at, over that time frame, energy intensity rose and fall a little bit with population. But then during the industrial transition to an industrial economy, energy intensity of the United Kingdom went up over 50 percent, over about 100 years and starting in the late 17 hundreds, and then it's been declining ever since then. So one interpretation of that is that once steam engines effectively and coal were combined to enable industrialization, resources were effectively so abundant that you didn't have to be that efficient at using them to grow the economy. So energy intensity increased and that's you know how much energy input is there divided by the economic output of the country. So energy consumption increased rapidly. GDP increased rapidly also during that industrial transition time, it just turns out that energy increased more rapidly than GDP did, and then since the late 18 hundreds energy intensity has been declining in the UK. So one could view that as a signal that resources were more scarce and that you had to start producing more output for less energy input and that after that time. So we're mostly used to energy intensity decreasing and thinking that as we decrease this number we're getting less dependence upon resources, that's not all of the story.

Chris Nelder: So if your economy started to enter into a recession, for example, but for some reason your resource consumption didn't fall as rapidly it would actually look the other way. It would look like your energy intensity was going up when in fact it's more like just an artifact of the math.

Carey King: Could be. Another way to think about it is, I suppose what economists call elasticity, or what's the sensitivity of one input to its output. So for example if you say what is the relative change in GDP compared to the relative change in energy consumption. So if you look at the worldwide data for energy, consumption and gross world product, for every 10 percent change in energy consumption there is a 6 percent change in GDP. So that also works in the in the opposite direction. So if I increase energy production I increase GDP. If I decrease energy production then the same trend works in the opposite direction. So the same kind of goes for energy intensity. The more I'm leveraging a large amount of GDP on a small amount of energy, I don't have to theoretically increase energy as much to grow GDP. But if I decrease energy just a little bit there's a lot more GDP leverage down that energy to go down even more quickly.

Chris Nelder: Right. So this is something that never really sat right with me about energy intensity, it's that GDP is the denominator. Because GDP, as explained in our episode 13 on peak oil, is really a very big gross metric that includes all sorts of societal goods and bads. For example if we go to war generally boost the GDP but not too many people would actually say that going to war is a good thing. So do you have any similar reservations about energy intensity as a metric?

Carey King: I suppose not more than what you might have just said there about GDP as a metric. You know is gross domestic product the wrong metric because it includes as you say bads as well as goods as increasing GDP. Things like war and destroying infrastructure and then rebuilding it and saying that is an increase in GDP. So, probably nothing more than to say that there are other alternative metrics such as genuine progress indicator or other things that that might more accurately indicate what we value about what the economy should be doing.

Chris Nelder: Right. OK. So the improving energy intensity ratios of the developed world is often cited as evidence that the developing world can go down the same path. That they could ultimately obtain a higher standard of living with a highly energy efficient economy and hopefully with a low enough carbon emissions to keep us below a 2 degree warming threshold while enjoying a relatively comfortable lifestyle. But there's also evidence that this is just an artifact of looking at the energy consumption and emissions of a given country in isolation. So, for example, the U.S. outsources much of its manufacturing capacity to China. And consequently the energy consumption and emissions of the U.S. went down. But China's went up. So if you look at the whole world together, the evidence for this decoupling thesis is quite a bit weaker. Have you looked into that question?

Carey King: I've thought about it a little bit but not too in depth as some other people. But the way to think about it, that that I enjoy is to think about it instead of the energy consumption and GDP that occurs within a geographic region we have to think more about the total resource consumption that occurs from the people in that geographic region relative to the GDP of that region, for example, so if we're importing a lot of resources from China as you're saying because they're doing a lot of manufacturing, our environmental or energy footprint, if you will, is based upon our consumption not as our production which is missing from the generic energy intensity number. So that's the main way I think about it. And the other one, as you're saying, is you know can every country be at the same energy intensity. Well countries are not equal on many metrics. And the question is whether whether developing countries that might have higher energy intensity or just a lower standard of living in general, will attain our standard of living and developed countries and effectively whether our economic system will let them do that. If your country in Africa you don't produce heavy machinery, have an education system to train people to produce heavy machinery, and manufacturing and hospitals, are you really going to leapfrog us by going to renewable energy infrastructure if you have to buy it from us and Asia rather than developing yourself. My thoughts at the moment are that you can't leapfrog the developed countries by just buying stuff from them. We're probably not going to sell it to you at a price that allows you to do that.

Chris Nelder: But in energy intensity terms is there any indication that on a global level energy intensity is improving?

Carey King: It has been decreasing. Although I think in the last 10 or 20 years I think maybe not as much. It's been relatively constant, and a lot of this I think was influenced by China joining the World Trade Organization and becoming essentially a global manufacturing hub and using coal mostly for that. So I think the trends of energy intensity have been declining basically for, certainly since World War II, globally are not decreasing as much anymore. Certainly as you say weight it has a market exchange rate basis, as you might say, instead of a power purchasing basis.

Chris Nelder: So you would, if I'm hearing you correctly, you are a little bit skeptical of this decoupling thesis as sort of a global path to decarbonizing the global economy.

Carey King: Well decoupling from energy and I would say it different than decoupling of carbon. So first...

Chris Nelder: Sure. Fair enough.

Carey King: decoupling of energy. So no, I don't see evidence that we have decoupled economic productivity or GDP from energy. If you just plot how much energy is being consumed per year, or really power, the rate of energy consumption versus GDP, it looks almost like a straight line. Right. More GDP equals more power consumption.

Chris Nelder: Just a one to one relationship.

Carey King: It's pretty close to one to one. Not at a slope of one but it's pretty close to 1 to 1. So. So that relationship is still holding. And as I said, it's also evidence to this elasticity calculation, which would say that every 10 percent increase in GDP is accompanied by six or seven percent increase in energy consumption. So then you translate that to carbon and, I mean just from the fact that most of energy consumption is still fossil fuel driven, we haven't decoupled from the carbon aspect as yet either. I think decoupling or producing electricity from mostly no carbon sources is a lot more feasible than, you know, things that aren't powered by electricity, things are just generally by heat, and use resources of heat and transportation are much harder, but that's a larger question. So the question of what it costs to transition to carbon is not obvious about how fast you can do it without essentially putting yourself into a recession by the nature of the transition itself. But that sort of a maybe a larger conversation or question we can address.

Chris Nelder: So one of the key metrics in this area of study is EROI, the energy return on investment, which tells you how much energy you get back for a given investment and an energy source. And we covered this topic at length with our mutual friend Dave Murphy in Episode 7. So I don't think we need to discuss the basics on that again. But one of the things that's always challenging about EROI, especially for those who aren't geeks like us are interested in it for intellectual reasons, is how to make it policy relevant. And this seems like an area where it really should be relevant. Is there a relationship between energy intensity ratio of an economy and the EROI of the fuels it burns?

Carey King: You mentioned the term energy intensity ratio, which is a term I use to compare the prices of fuels to energy intensity of the economy, it essentially scales in a similar way as energy return on investment. So essentially scales as, in this case, really, power produced per year divided by power input per year. So how to make EROI or net energy concepts policy relevant, I think, is to explain how they can and can't translate to economic metrics. So, effectively energy return on investment where you calculate the energy produced over the lifetime of the technology saved divided by the energy inputs over the lifetime of that technology is inversely related to cost. So EROI in this case has energy output in the numerator and the cost of energy as energy output in the numerator. So by definition they are inversely related. So one way to translate between the two is to simply plot them next to each other and sort of observe this inverse relationship. What they're essentially looking at two sides of the same coin. Or there are two ways of talking about the same issue. Cost is in some sense reflecting the technological capability of providing energy service and EROI is reflecting the same basic principle, but just in some sense another language. So I basically view EROI and cost as looking at the same concept, but they're just inversely related.

Chris Nelder: Okay. So if you have declining EROI as we have with fossil fuels generally speaking, can that tell us anything about the future trajectory of the economy?

Carey King: Well it certainly tells you how things have been going. So if we sort of take the idea that EROI and costs are inversely related, for example, the calculation that's actually done for things like the oil and gas industry that use annual economic numbers these are rates right, the rate of money per year or the rate of energy per year comparing it to power, so a power return ratio is inversely related to price, just like an energy return ratio is inversely related to costs. So when we're looking at the trends of what people are usually calling EROI it's usually a reflection of prices or the state of the economy at the time. And if you're thinking about prices in the state of the economy at the time these are a function of both sort of production and consumption. What are the ways in which people can consume energy and can they afford it? And what are the ways in which people can produce energy and can they afford to produce it? So a declining trend in EROI is effectively giving the same information as an increase in prices, and the question is is it too high or too low is there some threshold that's meaningful. And I think this is a little bit more well understood for thinking about expenditures on energy than it is the energy return on investment or the power return on investment for individual technologies. So I don't think I answered your question but...

Chris Nelder: I mean I take your point that if EROI is declining or increasing that people are going to look at that as a proxy for price because what people really think about is price. But EROI itself is priced blind. It's just about the energy. Right? So if you're looking at a declining EROI trend, let's say in an energy trend for fossil fuels, shouldn't that tell you that just on a purely energetic basis, never mind the price, that it's going to be harder to have economic growth in the future?

Carey King: Right. So let me address whether EROI is a price independent or an economy sort of independent metric. And my answer is it can be, but a lot of times when people talk about it it's not. So basically the way that EROI power returned ration can be independent as much as you can from economic issues such as prices or what people can afford is if you literally don't embed information in a calculation that involves money or effectively decisions by people. So as soon as you want to talk about something like well if I spent some money on engineering services to drill an oil well and then I use the money to convert to energy and that money as some proxy for the amount of energy that is consumed, I have just instantly there included information about the economy and to the calculation of EROI and is no longer independent of decision making and the state of the economy.

Chris Nelder: Point taken.

Carey King: So the only way to make it have it independent of the state of the economy is to do as you indicated which is effectively to only use power or energy information. So this is one thing I did in the paper from last year and the journal Energies, which was to take the data from the International Energy Agency, and they have a category called Energy Industry Owned Use. So this is effectively the energy carriers and potentially some primary energy resources consumed by the energy industry, to produce that energy. So it's effectively as pure of a energy rich on investment, but in this case I call it a power return ratio because it's energy per year units or energy per year that are spent, production or investment, and calculate this number for the data for about 44 foreign countries for as much as existed in the data until 2010, and that number showed that for the last say 20 years or so that this number has been between effectively 15 and 19. But before 1983 this number was probably larger than 30. So just using this attempted calculation of energy return on investment that is most independent of economic information, there still seems to be a decline from essentially before the 80s and after the 80s and that has been relatively constant since then. So effectively that's a global metric. So effectively we start from there and then include other economic information at that point you are starting to understand as soon as you include other information in there it's effectively economic proxy information I guess.

Chris Nelder: Right. Just to be clear that's an average of energy investment across the global energy sector.

Carey King: That's correct. So this is a calculation for every country in the list and then each country was averaged based upon how much energy it produced. The global average number so. The global average is around 14 units of gross energy production and their primary energy production for every one unit of energy carrier input.

Chris Nelder: Down from 30 prior to the 80s.

Carey King: Yes something probably over 30.

Chris Nelder: Okay. And I just wanted to make another clarification here. You have made a distinction a couple of times now between power and energy and I want to make sure I understand that. So when you say power, are you talking about sort of final energy or maybe some would call it exergy as opposed to the energy input of the original fuel?

Carey King: Right so if you're standing the difference between power and energy is actually pretty important and key, and I think it's left out of a lot of conversation. So when I say power I literally mean if you're using units of power which are units of energy over time. So for example a watt that is a unit of power, it's joules per second. And energy, a unit of energy is joules. So the way to think about it from a EROI standpoint is say comparing what is usually done as a calculation of energy return on investment for a wind turbine to the calculation that is usually done for the oil and gas industry. So for a wind turbine, you would say OK how much would it cost to manufacture the wind turbine in energy terms and install it at the beginning and then some maintenance over say 30 years and then potentially decommissioned it in 30 years. And there's some energy requirement today, over 30 years and then again 30 years later. So it's a power consumption each year, it's an energy consumption over power consumption over time. So essentially power times time. I added up all the energy that it took to produce and operate the energy, this wind turbine over a span of time of the lifetime of that wind turbine. So it's a quantity of energy. I summed all of that energy over time literally integrated power over time to come up with units of energy. Similarly you would say this is how much power it produces every year and I've integrated the power it produces over a year over each year over time to produce a unit of energy. So that energy return on investment compares to cost. If I thought about what the cost of electricity from wind turbine I would also think about all the money spent over the entire lifetime of the wind turbine and compare that to all the energy generated from the wind turbine over its entire lifetime. So that's an energy metric divided by a cost metric, how much money is spent over an entire lifetime. So the difference let's just say, which is normally done with calculating EROI if you will of the oil and gas industry is the data used are literally energy production per year and energy inputs per year, because they're usually using data that are annual data. So if you're using oil and natural gas production, this is a unit of power, and they're calculating this every year usually or every five years or something like this because the data are annual, and if I even use economic data as proxies for energy consumption the economic data are dollars per year usually, right. GDP is is not a fixed number, it's a rate. GDP itself is a dollar per year number. So there are both rates. So if I'm going to use the rate of energy production divided by the rate of energy input then it's really literally a power ratio or a ratio of power divided by power. And that was more closely relates to price, because prices are a reflection of what's happening now and the rate at which things are happening. Can people afford to consume now, or afford to produce now? That's what affects prices, not the cost of installing a technology now, it's the prices in markets and other things.

Chris Nelder: All right well so then let's talk about the price of fuels themselves. I mean after all oh anybody ever seems to care about is the price, particularly as we discussed in episode 13 where peak oil is concerned. What are your thoughts about that? Can EROI trends actually tell us something useful about the terms of oil prices?

Carey King: Yeah. So I think they can for effectively the same standpoint that they're looking at the same issue, EROI is looking at the same issue as prices just from a different lens. Effectively because the concepts of net energy stem more from say system energy analysis or ecological systems analysis in which the concept of limit or the concept of feedbacks is usually inherent and their trying to understand the system. It is a more natural way for people with that background to think about what the feedback might be, what EROI is too low, in which the system configuration in the case of ecosystems would have to change because people studying engineering systems often or ecosystems are thinking about OK what are the constraints on how the system operates, and let me try to understand how to design the system or at least understand it within the constraints of that system. Economists are not normally trained and usually don't think about some sort of inherent constraint on the economies. So the idea of thinking about a threshold in prices as creating some sort of fundamental constraint being a reflection or a feedback of some sort of fundamental constraint on the economy are probably not as common. Now, short term the answer is yes. I mean they understand that people change consumption habits based upon short term prices. Right. So this is commonly known. But longer term it's less clear that these kinds of ideas are taken into account and say long term macroeconomic modeling. In terms of thinking about net energy or EROI trends versus the trends of prices. Because I've been working a lot trying to understand how to translate that energy metrics to economic metrics, I've come to understanding that what's important from an economic standpoint is expenditures and not prices. So what I mean by that is let's just say I'm purchasing oil at some price, but the question is how much oil did I purchase, and that turns into my expenditures relative to the GDP of an economy. Now this creates a relative metric to know if something is too high or too low. And when you look at the data on say cost share of energy, essentially how much the world spends on energy, whenever it's below 6 percent, the economy has traditionally let's just say since WWII has been kind of humming along, that energy and this sense is relatively cheap. When it's above say seven or eight percent then energy looks expensive. So price by itself doesn't tell you necessarily whether energy is expensive or not. The expenditure is relative to GDP or relative to incomes of consumers, that is what tells you whether it's expensive or not. And so if this ratio gets you high, that would be a metric of whether it's expensive. To give you a scale of expensive versus not expensive energy in terms of cost share or what expenditures are relative to GDP, take England and the United Kingdom as an example of a long term time series. So before the industrial transition, say before the late 1700s, over 30 percent and sometimes 40 or 45 percent relative to GDP were the energy expenditure. So if you took all the energy expenditures in England and divided advice GDP would be 35 to 45 percent. So that's preindustrial, pre fossil fuels, and now that number is less than 10 percent. So we can in some sense clearly say all right, the energy was expensive preindustrial and it's cheap now.

Chris Nelder: OK, well that makes sense. And I think you've also looked at the same question from a food standpoint. So total expenditures on energy and food price time consumption relative to GDP as a percentage of personal income and using that as a way of measuring the real cost of energy. So can that tell us anything about where we're going?

Carey King: All right so you bring up an important point, which is effectively how to think about food, and it's often something people think about sort of from a hierarchy of needs standpoint, sort of food, water shelter or throwing energy in there, kind of as a hierarchy of needs or things high on a list of needs. So preindustrial food and biomass were the energy fuels, our resources, for much of providing energy services are physical work. So humans would be food to provide physical labor on the farm and do actual physical work that tractors and diesel fuel does now, and a lot of the biomass was used to feed animals to provide this physical labor on the farm preindustrial. So you clearly have to include food and agricultural biomass as part of the energy supply from a pre-industrial standpoint. So the question is should you think about it now? And in my mind there's no reason not to think about the effectively biological needs of humans to eat food as a core expenditure that's necessary to keep the economy going because people are part of the economy. So even though we don't use food for direct energy services as much any more in developed economies, for example, I eat food as I sit here and type on a computer so I'm not doing physical labor, I'm still a core resource. So in my mind if we want to understand the core energy expenditures over time I throw food in there, and I want to track these over time as much as possible. And if you track things over time, globally it looks like around the year 2000 was the cheapest energy in food effectively in the history of mankind. So energy and food, primarily food, were getting cheaper relative to GDP since industrialization until about the year 2000, and since then food has not gotten more expensive but it also hasn't really got cheaper and energy has become slightly more expensive since that time. So it appears as though we have reached a low point in energy and food expenditures relative to GDP or what I would use as a sort of fundamental metric of cost.

Chris Nelder: Yeah and so you know I think a skeptic would listen to all this chatter we're having here about EROI and costs and so on and say well you know what's the big deal? I mean OK so EROI has been slowly declining, the cost of energy and food has been following steadily for over a century since the dawn of the industrial revolution. And that's often cited as evidence that for example the warnings of you know the likes of the Club of Rome and limits to growth or Paul Ehrlich's population bomb or even more recently the fear of reaching peak oil production, but those were all just false alarms. I mean after all we went through a year of oil prices under 60 bucks when just three years ago nobody thought that was even remotely likely to happen. So since the prices of all commodities have been falling for 30 years and we're now looking at a worldwide glut of most commodities rather than the scarcity everyone expected, should we not conclude that we've just banished the specter of scarcity entirely through efficiency and ingenuity and there is no reason to think that we can't continue to stay ahead of scarcity fears?

Carey King: So I get argument to that at a couple of things. One is sort of addressing that question or thinking about this question is one of the reasons why I wanted to look at sort of the longest time series I could at what the cost of energy and food were relative to GDP. And since those costs put together are no longer declining. In my mind that's evidence that scarcity and in some sense or at least the finite size of the earth is starting to impose itself on the economy and limiting some options. So over the last 30 years is simply not long enough of a time series to put it into perspective. So that's why I kind of look at longer time series and try to say what was it like before the industrial transition? What is it now well after the industrial transition? So the interesting thing about thinking about energy and food costs is that when I ask people this like can they be zero percent? Can I spend zero percent of GDP or my income on food and energy? Most people say no. Right, so if it was a high percentage 200 years ago and it's a much lower percentage now, that's no longer getting lower and it can't get to zero, the you have to ask yourself OK what is the lowest number that you can achieve, and how do you know if you've achieved it? And right now it's not like technology has stopped. I stated 2000 was the low point in food and energy costs. But since then we've had a tremendous increase in wind power instillations, solar power installations and you know hydraulic fracturing and horizontal drilling for tight sands and tight formations. So you know technology didn't stop in the energy sector but it also didn't make energy cheaper than it was back then. So just looking at the data it tells me something's happening independently of whether we think technology is improving which it clearly is. A second way to look at it is just prices again as a reflection. But the BP data on you know world oil prices is instructive here. Effectively the beginnings, the dawns of the oil age before 1900, the average oil price was say around $50 a barrel. And then sort of the late 1880s all the way until 1973 before the Arab oil embargo, oil essentially averaged $20 a barrel. I'm referencing real dollars here, so everything that would be inflation adjusted to say 2015 dollars. But since 1973, so from 1974 to today, oil's averaged again over $50 a barrel. So that was a time in which it was expensive or relatively high price before we had gotten used to basically using oil. Then there was a long period of time almost 100 years where it was relatively cheap around 20. And now for the last 40 or 50 it's been averaging 50, so I wouldn't say that oil is still as cheap as it used to be from a price standpoint, I would say it's expensive as it was at the beginning of the oil age. Now the difference is you know why can we effectively quote unquote let's say afford $50 a barrel oil now or $60 a barrel oil? Well that's because we don't consume as much of the same service. The response to scarcity in the 70s was literally to change the way we consume energy. We did a few things. One is look for oil in places and go get it places where we haven't gone before such as North Sea, the Gulf of Mexico, Alaska. But the other response was to say let me change the other side of the equation instead of supply when we changed the demand side part of it and let's become more efficient effectively enacting efficiency standards for the first time. So we can't really look at the 1970s as not being a fundamental indicator of scarcity. They were, and the events of the 1970s were were in fact fundamental indications of scarcity, and there were fundamental responses to those implications that will never go away. I mean we thought of efficiency effectively for the first time after that. That's a fundamental change in the way that the economy has been organized.

Chris Nelder: Well I'm glad that you mentioned this issue of price over long periods of time because it's a really hard question, it's something I found particularly bedevilling especially when you're trying to look at the relationship between resource consumption and economy over time between different cultures or different times, it's just really hard to find a good metric with which to make the comparison. For example, how do you value a loaf of bread or a kilogram of coal in the 17th century compared to the value of those things today when the money itself has changed and morphed and been debased over the centuries? Or what's the right unit of account for comparing the value of things across cultures and centuries. I mean even gold doesn't work for that because it's been hoarded and rationed and debased from time to time and place to place. You know I was having this discussion with a hedge fund manager friend of mine who suggested that the services of a prostitute might be a good unit of account and measure because they have pretty much the same value to all people in all places and times whether it's Rome in the year 300 or the U.S. in 2016. So how do you think it's best to compare the value of things across these long time spans and places?

Carey King: Well I think your hedge fund manager friend might have a point there whether the prostitutes prices are more relevant than the cost of energy and food. But since I study energy and food and i'll have to discuss about how those things relate. So I'd be interested in hearing any definitive literature is on the other subject. Yeah so as discussed earlier, this is one of the fundamental reasons why I thought about looking at the cost of energy and food, and basically as expenditures over time. Because ultimately we normally calculate, if were talking about net energy issues the energy return on investment of this technology which is that technology and that technology, ultimately we're making a choice about how much of each technology we want to have for one reason or another whether it's cheapest cost or highest EROI or what have you, but we don't just use one, we use a mix of them. And this is for a variety of reasons that relate to costs and some of which don't relate directly to costs or net energy returns. So effectively we have a set of technologies and we have to find ways to aggregate them together to add them up for the whole economy, and the total expenditures on energy is effectively that aggregation done by the collective decisions of everyone as well as the regulations and market principles or lack thereof that are put into place. That's why I look at energy and food expenditures as sort of the ultimate society wide metric of how much food energy costs over time. Just like the power return on investment is inversely related to price, effectively right now I think the best metric of the power return on investment of society overall is essentially the inverse of the cost share of energy. So if my energy expenditures are 10 percent of GDP and if I invert that I get 10, and so that would be the rough proxy for what the net energy of the world is. My research is trying to compare these metrics and try to understand effectively how I would aggregate a set of individual technologies into the overall system. And you have to have ways to translate back and forth. It's not only a question for net energy analysis, it's a question for macroeconomics in general, tying to microeconomics and they're fundamentally not very well linked either.

Chris Nelder: Right, so you're basically sidestepping the problem of converting the equivalence of currency across these different terms and cultures by just comparing percentages, essentially.

Carey King: Right. So I think looking at in this case relative metrics or the percent of some allocation that is used for purposes X, Y or Z ,in this case the percent of all money that's used for energy and food, it effectively to me is such a defining metric of our society that it simply can't be ignored. If we include food expenditures and energy total food expenditures for work in preindustrial England, then aside from the non-food energy expenditures if we add what we're spending on food for physical labor then the number gets to 70 or 80 percent of GDP was spent on energy. So basically I mean we can view the modern economy or industrialization as declining that ratio is working to make it so that people don't have to work on the farm, because farming was something you have to do, right, the hierarchy of needs. You have to have food otherwise you aren't going to survive. So technology's effectively and minimizing the number of people working in the quote unquote energy and food sectors over time. How far can that trend go? What's the minimum number of people we can hire in the food and energy sectors is sort of the same question in a different way, a different metric of saying what is the percentage of GDP that is spent on energy.

Chris Nelder: Yes so you found this interesting little threshold you know around 6 percent, 6 to 7 percent. That's where we tip from you know the economy humming along to slowing down.

Carey King: I wouldn't say I'm the only one who has found that.

Chris Nelder: I was going to say, I think James Hamilton came up with that same number, didn't he?

Carey King: I've read some of his papers in which he looks at effectively the rate of change of prices, usually oil prices and it's relationship to recessions and I think this off quoted metric is a 10 of the last 11 recessions in the U.S. were preceded by "spikes in oil prices". So the short term phenomenon is there. So having high expenditures on energy effectively is about the rate of change, like is the rate of change too fast for people to adjust, and people are adjusting could be things like getting more efficient appliances or more efficient cars. So things can change too quickly to adjust. There can also be reasons you can't adjust which is maybe you can't afford it either so. So I think one of the questions going forward is trying to understand that kind of relationship and it relates to thinking about an energy transition. So let's just go back to the rate of change, or the 6 or 7 percent threshold kind of number, there are not a lot of data points for this theory. You sort of look at it and say there's a correlation and then probably can't address causation. I think the energy and economy are so fundamentally linked that the causation question is sort of a side issue that's not that relevant for investigating. But the only two times that expenditures are more than 8 percent of GDP was the late 1970s after the Iranian revolution and

Chris Nelder: And the Arab oil embargo.

Carey King: Which occurred previously in the 70s, and then cut off of you know large oil supply I think around 8 million barrels in 1979. So there's that time. And then there's 2008. And those are the two times when it was 8 percent or higher. And both of those times were a time of great economic uncertainty and wondering if the economy is structured in the right way, wondering whether people understand how the economy operates. So obviously there was a rapid change that was you know in some sense unpredictable exactly when it would occur. In 1979 the rate of change leading up to 2008 was slower, but still we'd say relatively rapid in terms of people adjusting to it. I mean there's just no competition for resources from China becoming our sort of manufacturer of the world in this kind of thing was still relatively rapid. And so it culminated in that high energy price. And so those are the two data points. The economy doesn't grow and the economy grows below them. So those are the two points we have. So it remains to be seen how much it holds. But I lend a lot of credence to the idea because I understand that you know if feedback can be non-linear, right, a 1 percent change from 4 to 5 percent of GDP spending on energy is not the same as a 1 percent change going from 7 to 8 percent. And that's just born in the data.

Chris Nelder: Interesting. Well OK so on that point you know we've had two and a half years now of low oil prices and that has forced many analysts to rethink their views on the global outlook for oil and the economy. For example I thought we'd hit that global production peak when prices were high but now I think maybe the global oil production peak might have actually been last year 2015, although it's a bit too early to say, when prices were low. And there's still a pretty vigorous discussion about how oil prices can affect the economy, like whether low oil prices will eventually starve supply and set up a price spike in the next few years which will really hurt economic growth, or whether the global economy can actually tolerate high prices just how high that number can go before it really puts the brake on economic activity. And I hear you know injecting this concept of the rate of change as well, so what are your thoughts about that, about you know where we're going in terms of the investment in the oil sector being sort of starved right now with this low price environment and what that might mean for not only the future of oil prices but the future of economic growth globally?

Carey King: Right. Yeah because I'm in some sense to try and emphasizing the difference between energy and power, the real emphasis there is about how the rate of change effectively. So, you know in the 2000s before the great recession people were often discussing you know whether oil would reach a price of $200 a barrel.

Chris Nelder: That's right.

Carey King: And you know what would happen or what kind of alternatives we would have this kind of thing. Well what a lot of people didn't know then, and I'd say if you're paying attention you should surely know now, is that the economy can't have $200 a barrel oil. And that's because of this feedback between the expenditures relative to GDP. If you had oil is a $200 a barrel the oil itself would cost close to 8 percent of GDP and the economy simply is not configured to spend that much on energy. It has to essentially restructure itself in other ways, other things have to change, other parts of the economy have to change to allow for that level of expenditure. And the way that the economy restructures before it gets to $200 a barrel is to go into recession. So that's kind of the restructure mode. And if the rate of change of demand driving prices high or rate of change and a drop of supply also makes prices high, if this rate of change is too quick then this is reflected in prices and therefore because the rate of change is too quick people can't adjust, and that much makes expenditures go up because they don't change the quantity of they consume fast enough. So recession will make people change the quantity they consume more quickly by firing people and depressing economic activity. So you literally don't have to consume as many resources. And so this is sort of a fundamental question that I've been wondering about. Because there's a you know a tenet of net energy that is espoused by say you know Joseph Tainter that net energy of a society relates to its structure, right. Or the idea that this input output relationship how much energy out do I get for energy input, does this actually relate to the structure of the economy. And at this point my conclusion is a single number of the energy return on investment of society or the expenditures on energy as a percentage of GDP, there's not just one structure that can relate to these numbers, there could be multiple structures. I could have a highly unequal society at one EROI or an equal society at another. So I recently produced a paper that was trying to essentially put this theory to the test or essentially see if I could see changes in the structure of the U.S. economy if you will, essentially as a function of energy and food costs, or a net energy concept. And so that's what I put into the journal Biophysical Economics and Resource Quality, which is a new journal by Springer to help people explore some of these relationships between economic and resource issues. And effectively there were two time periods in which there were I would say fundamental structural changes in the economy as measured effectively by the input output tables of the economy. Essentially how much money as each sector of the economy purchasing and selling to each other sector of the economy. Did the flow of this money change in that structure? And one turning point of structural change was the late 60s and the early 70s. And another structural change was around year 2000 when energy and food costs had reached their bottom and then they started increasing from there. So there were two time periods at which there were fundamental questions about energy and at which the relationship of the economy with energy changed, and these are essentially fully reflected in the structural changes of how money flowed throughout the economy overall.

Chris Nelder: And if I'm not incorrect the paper you're referring to there is "Information Theory to Assess Relations Between Energy and the Structure of the U.S. Economy Over Time" .

Carey King: That's correct.

Chris Nelder: OK. We've got that one in the show notes for all of our studious little listeners out there who want to go check that out. OK so since you're one of the few analysts who's really studied the EROI of various fuels including renewables and fossil fuels, I'd like to wrap this up with hearing your thoughts about the project of energy transition over the coming decades. I mean recognizing that the study of EROI or I guess maybe these days it's more probably the field of biophysical economics or is it life lifecycle analysis, LCA, these days? That's all still pretty young as a field of study, and that important foundational work still remains to be done. And deriving methods and data for effectively comparing the EROI of fossil fuels and renewable technologies on an equal footing over time, and recognizing that renewables are a very fast moving thing to try to study with cost and capacity factors improving significantly like on an annual basis. So do you think this area of inquiry of LCA can really usefully inform us about what energy transition can and cannot do or what paths we should take or not take?

Carey King: Yeah, so that's kind of a good and sort of one of the ultimate questions. So one of the drivers of my research thinking is that the lure of net energy analysis, which was effectively started in response to the oil embargo in the 70s, was to understand more about fundamental relationships of energy in the economy. You know the lure of energy return on investment and these kinds of concepts is that they seem pretty intuitive. You think OK yeah I've got to get some extra energy out of the part of my economy or ecosystem that is effectively tasked with giving energy to every other part of the system. And you think OK well that seems like an intuitive notion. And then the question is OK, Oh now I have to quantify it and relate it to something in order to make a decision. The concept is neat but the quantification is needed. So that's why I focus on trying to understand how to relate these concepts to the economy, to try to understand what data and prices or what people would consider normal economic data. How can I phrase them in a way that relates to net energy concepts. And I've come for a long enough in my thinking to realize that you have to include things that are not inherently included in net energy analysis or things that are fundamentally difficult. And so these are things like wages and profits, debt and employment. So these are things that economists would think about and want to model, and biophysical economics person might pay more attention to material and energy flows and they might also pay attention to how many people are employed and doing one activity or another. But the concept of debt is not something that's sort of inherent in thinking about say traditionally ecosystems or engineering systems. So what I'm currently working on is effectively some modeling that incorporates the best aspects of both ways of thinking, right. A biophysical way of thinking, where there is sort of an inherent limit to something in terms of the space in which the economy can reside. So like something that represents that the earth is finite as a potential feedback. I think this is important to put in because we know the earth is finite, and so the question is is the finiteness feeding back to us now? How would we know? Well if you don't put the concept in your model, you can't answer the question so, so that's important to put in. From the economic standpoint, the ideas of debt and wages and profits are are inherent in the economy and things that we have to deal with. A lot of macro economic models don't even consider the idea of debt or money, right, so a lot of the economic models that do projections for a low carbon transition don't factor in debt as a possible influence on the outcome. So that's kind of silly because this is obviously focus of a lot of discussion and we need to think about the concepts of debt. So integrating a few fundamental pieces together in a fully coherent model even if it's quite simplistic is I think a necessary goal at this point and something that I'm working on it is necessary to indicate larger policy implications of an energy transition. So these things are resources extraction, wages, profits, what's the interplay between those, debt, population growth and employment effectively. What percentage of the population is working. I mean these are all issues that are affecting us right now and that economists are struggling to explain. We're in a period of high debt, lowest interest rates in the history of central banking. according to me we passed the lowest point of energy and food costs in the history of mankind. So I think these these things are occurring relatively simultaneously. And these are major trends in the world. Population growth again is slowing down, still growing but it's slowing down, and the demographic transition of aging population in developed economies is a fact of things happening. So if you avoid these fundamental factors and try to explain the world I think you're leaving a major one out. So I'm trying to include these major factors in a single coherent model to have something to say which could be resource constraints and net energy gains do have something to say about why interest rates are low as a reaction and debt rates are high as effectively a past assumption that future growth would be higher than it turned out to be.

Chris Nelder: Ok, but do you think all these things can actually tell us for example if energy transition is even possible, or if it's even a good idea?

Carey King: All right, so if I'm going to do an energy transition the key factors are, so energy return on investment is one way to look at it, right. To say the rate at which I transition basically means the faster I do the transition the more materials, the more energy, the more people are going to be mobilized to that activity. And as we've discussed in the podcast so far industrialization has effectively been moving materials, energy and people as a percentage of total mobilization out of the food and energy sectors. So a transition to a low carbon energy is effectively a reversal of that trend. Particularly because we've reached the low point in food and energy costs. And oil and gas boom is effectively a microcosm of say a low carbon energy transition. There was a tremendous amount of oil and gas drilling after the great recession in 2011 to 2014. It was such high drilling activity that the drilling rates, the rates of which people have to pay to higher rigs to do the drilling went up because there's a certain supply of drilling rigs. So at least some level of scarcity, right. How many drilling rigs are out there. So this drove up the prices for those drilling rigs and necessitated a higher price for oil. If you're not drilling as much there's not as much pressure on drilling rigs and therefore the price of hiring in drilling rig goes down. So that's responsible for over half of the drop in cost of producing oil is just a slower rate of drilling. So the same thing would happen for an energy transition. And except it would probably be one or two orders of magnitude more intense than the oil and gas boom. It would involve a larger section of the economy than oil and gas and it would have to occur across a larger segment of the industry. So energy return of investments has something to say about that in terms of what does the energy return investment of the technologies that we're using to essentially transition away from other technologies. If we're using lower EROI technologies to transfer away from higher EROI systems than we would have to assume a higher mobilization than would otherwise occur if the energy return investment of the systems were higher than the systems we're replacing. The difference in capital and operating costs has a lot to say about that as well. Those ideas are kind of hidden and embedded in energy return on investment calculations. They are effectively hidden and embedded in cost calculations as well, but most of the low carbon systems people are talking about are much more capital intensive or capital heavy as a percentage of total investment. And that effectively is a driver for making the transition more difficult in terms of having a lot more money up front and putting pressure on the economy than otherwise.

Chris Nelder: So trying to put all that together in terms of giving us some guidance about energy transition, I mean are you saying that whether we continue to transition toward renewables or whether we continue to consume fossil fuels that the cost of energy expenditures are going to go up no matter what? Or does this data give us any guidance as to which is the better path?

Carey King: Right, it doesn't say that the cost of expenditures will go up no matter what, but it does imply that the rate at which you invest in the energy sector in general let's say effectively makes prices higher. A higher rate of investment means a higher rate of employment, a higher mobilization of activity that go to the energy sector, so that you shift the size of the energy sector making it bigger relative to the non energy sector. So effectively fewer people in a non energy sector are paying the salaries of an increasing quantity of people in the energy sector. Even though it's still a relatively small percentage, just that shift puts pressure on prices. So the faster you transition the more that shift occurs and the higher you would expect the price of energy to go during the transition. So the faster you transition the higher the price is going to be in the short term. And we know that high prices in the short term, if they occur quickly, don't allow people to adjust consumption, which makes expenditures go up, and then we know if expenditures are high enough you can put the economy into recession. So that's one of the key probably policy questions and outcomes to answer is here's how fast you can transition it without putting the economy into recession.

Chris Nelder: Now that's a really interesting insight, but I don't know, I hear that and I just think yeah but the rate at which we do energy transition is no longer something that's really that optional, right? Or over which we have a lot of discretionary decision making. I mean we are fundamentally driven toward energy transition by the problem of climate change. I mean that's the fundamental underneath all of this. And the data that we're getting about climate change tells us that we need to be largely decarbonized globally by 2050, which in terms of the speed of energy transitions historically might as well be tomorrow. So bearing that in mind that we do have a sword hanging over our head here, and we have a deadline, what can this research tell us about how we should be proceeding?

Carey King: All right so in terms of the carbon mitigation question, I think that's the question. In my mind or my best guess and certainly my hypothesis at the moment would be that we would not transition to a low carbon let's just say less than 10 percent greenhouse gas emissions relative to the turn of the century by 2050 without putting ourselves and putting the economy into recession by that time. It's not the same as saying it's not something that is worthy of doing. But I think that's the tradeoff. Now the problem is that the economic models that are used to inform low carbon transition simply can't account for the rate of the transition. So the answer from all the models is that essentially they can't tell the difference between a low carbon economy and a high carbon economy in terms of GDP, right. The results from the IPCC reports that are put in there from the economic modeling effectively is saying if you transition to effectively zero net carbon economy by the end of the century to reach 2 degrees target that the percentage drop in GDP by the end of the century would be about 5 percent. Effectively their assumed economic growth by that point is that we would be say three or eight times richer by the end of the century. So they're like OK I'm going to be 800 percent or 300 percent richer if we don't do climate mitigation, but if I shift resources to do climate mitigation I'll be about 10 percent less, so instead of 300 percent to 800 percent richer, I'll be 270 - 720 percent richer. Effectively that saying I can't tell the difference. And the reason I can't tell the difference is because the rate of the transition isn't included in the modeling, and debt isn't included in the modeling. So in my mind, to understand the rate of transition I simply have to account for these things, and energy return on investment helps these ideas help us think about what the levels of debt would be if you don't have the resources to shift and you're shifting the payment into the future then that would increase debt.

Chris Nelder: Wow. That's a super interesting insight, that debt is going to be a major feature of energy transition.

Carey King: Right. So we have effectively increased debt, and in some sense one of the major differences between say today and late 70s when oil prices were high, let's just say 2008 and the Great Recession commodity oil prices were high, while in the late 70s the interest rates were raised to try to prevent inflation and solve the problem of stagflation. So they were basically the highest in history of central banking. Now we have interest rates at the lowest point in central banking. So for 20 or 30 years from the late 70s until around 2000 or the mid-2000s when interest rates eventually got drove to zero after the Great Recession, you could effectively have a business and just not actually fundamentally make a profit but just acquire debt and refinance along the way. Right. This was what people meant by "the Greenspan put", that he said the economy is not growing I'm just going to lower interest rates. So I'm thinking well I acquired some debt in order to have my business operate, I didn't quite make as much money as I needed to pay off the debt, so I'll just refinance at a lower interest rate. OK I'll do that again, and I'll do that again. Well you can't do it again now. Right. We're at zero. And central banks are afraid to go much below zero even though there is out there at a below zero interest rate. There is savings at a below interest rate. Because the thought is that people will just take the money out banks and put the cash under their mattress. So you have a backstop there so you can have tremendous negative interest rates. So this is the conundrum that's on. So try to understand the potential for people to invest in energy transition independent of the ability for people to invest in any profit making activity I think is unwise. I think we have to have it in net context. And right now the low interest rate and high debt situation is indicative of the difficulty of people finding profitable investments in general, but energy since it's a necessity has to keep getting invested in. You can't not invest in energy. If you don't invest enough the prices will go up, you will reach expenditures that reaches seven or eight percent threshold, and the response will be invest in the energy sector to try to solve the problem. In some sense, that's what we saw. The investment after the Great Recession was hydraulic fracturing horizontal drilling for oil production. That was the response.

Chris Nelder: Yeah right. And now we have some very attractive and profitable investments available in renewals, which is handy. It's going to take I think very patient capital in the long run. We're going to need substantial amounts of capital that are willing to accept 3-4 percent returns probably. But the opportunity is there.

Carey King: Interest rates are low, indicating people are having to say quote-un-quote a hard time finding investments like they're used to say before you know the 2000s or this kind of thing. I think exactly as you just said, people will have to become more used to investing in lower rate of return things, and the renewables are there, right. I mean you can make a profit investing in say wind and solar projects, but you just might not make as much return as you were investing in Apple or electronics 20 years ago.

Chris Nelder: Right.

Carey King: So that there is a feedback there that the more you have high capital infrastructure in the energy sector, in this case renewables like wind and solar are high proportion of the money goes to capital investing versus operating costs, this is a feedback towards a lower growth situation in the sense that you're investing in things that produce slower returns. You literally produce them back slower because they're paying back the capital of a long period of time. And that's not a horrible thing. That's just the nature of it. I think it's all part of the evolution of the economy in general. The longer the interest rate situation stays low the more attractive renewables look. Lower interest rates make renewals look attractive, because they pay back slower. So it's reasonable to expect that we'll continue to promote renewables going forward from now.

Chris Nelder: That's a nice thought. Well thank you Carey. This has really been an informative discussion. I appreciate you taking the time to come on the show.

Carey King: Thank you very much Chris. Thanks for having me.