“The determinants of economic health are timeless and universal” – Ray Dalio, How the Economic Machine Works
Over the long-term productivity is what matters most for economic growth (and development).
Traditionally, we tended to look at production as a function of how efficient (productive) labour and capital are employed – this relationship has been the bedrock of our understanding of economic development and growth for centuries.
However, by seeing production as the output of only labour and capital, we risk ignoring a fundamental component of development and growth: energy; more specifically, energy efficiency.
Prior to the industrial revolution, economies around the world functioned on ‘above land energy’ – solar (photosynthesis that helped grow food and trees for wood supplies), wind (which enabled maritime transport) and water (e.g. through irrigation systems and water mills).
Things have changed with the adaptations made to the steam engine by Thomas Savery, Thomas Newcomen and James Watt throughout the 17th and 18th centuries.
Their combined breakthroughs eventually enabled countries to tap into fossil fuels (such as coal, oil and natural gas) on unprecedented scales. This led to incredible technological innovations and advancements that finally freed a large proportion of people, on a global level, from the life’s necessities.
Fossil fuels represent energy from ancient photosynthesis stored over millions of years, sometimes exceeding 650 million years. Therefore, once consumed, it is impossible to be regenerated during a human lifetime.
By looking at changes in GDP over timeframes that do not account for this massive change in production and prosperity, capital allocators concerned with the long-term destiny of the world risk building an incomplete picture of the main drivers of past, current and future economic development and growth.
That is not to say that we should account for thousands of years of economic change – but only for a period that goes back long enough into the past to capture the pre-industrial revolution world (for example 100-150 years before), while also capturing the period since then. In this way, we account for the transition between what we can call ‘energy cycles’, enabling us to appreciate the importance of energy in driving economic development and growth.
Just as there are debt cycles, which result from debt levels growing to a point that overburdens the economy, there are also energy cycles. The chart below illustrates this.
During each of these energy cycles, people make use of certain energy sources in the most efficient way possible. This energy harvesting happens up to the point at which using that energy source is no longer productive – at this point economic development stagnates and consequently, economic growth also comes to a halt. Indeed, if we look at history, we see that there can be prolonged periods of very little-to-no economic development and therefore, mute economic growth.
Seen in this context, the period of economic prosperity achieved over the past 200 years or so, especially since the end of WWII is truly unique in human history and we should not take it for granted – rather, we should look to understand what has made it possible.
A key ingredient has been our capacity to put to work dense, cheap and finite forms of energy – fossil fuels.
Not accounting for this factor, while also supporting with more and more debt a socio-economic system that seems to value short-term consumption over long-term investment can have negative consequences for the future prosperity of the world and consequently, for investment returns.
What determines economic growth?
Before we try and answer this we need to make the difference between economic development and economic growth.
These two processes, in our thinking, are different in what they are but overlap in how they manifest. We will discuss this in more detail in a future paper. For now, suffice to say that economic development requires innovation.
Meanwhile, innovation requires a certain mix of cultural values, legal / regulatory system flexibility and access to resources which enables creative people to a) imagine new products / services, b) put their imagination to work (i.e. execute their ideas), c) bring their final work to market (should they be successful in executing their ideas) and d) be remunerated for the risks taken in creating something new. Where economic growth and development overlap is in the final two parts (c) and (d).
Growth means to consume something that was produced – the more consumption there is, the more demand there is and therefore, the more production there is. However, growth, in the absence of innovation, which is the key driver of improving productivity, reaches a point where it becomes an unproductive endeavour. In other words, consumption, which is the remuneration of producers, no longer is enough to justify itself and the processes of production need to be improved or new ones developed – this is done through innovation, i.e. the core element of economic development.
Now, back to economic growth…
There are several factors that determine economic growth but, at a big picture level, economic growth, or GDP, is a function of an economy’s workforce and the productivity of its workers. As such, GDP equals the number of workers times the output per worker. Transformed into growth rates this equation becomes: ∆GDP = ∆Workforce + ∆Productivity.
Over the long-term productivity is what matters most for economic growth. Productivity is a measure of how cost-effective one produces something – a key way to quantitatively look at productivity is as output per hour worked. Some examples below.
An individual (or a company, a sector, a country) can increase their productivity by either working more or working smarter (i.e. more efficient). Working more simply means spending more of your time doing your job (or jobs). Working smarter however entails investment into researching and developing new technologies, methods and production processes, i.e. it requires innovation.
Ray Dalio, in his Economic Principles, breaks down productivity into three main and interconnected factors: a) the country’s cost-effectiveness of labour, b) the country’s investment level and c) cultural values. To this list, I would add a fourth ingredient – exergy; this can be viewed as the capacity of energy to do physical work.
Energy and economic productivity
I was first introduced to the field of exergy economics while working at Woodford Investment Management, where I had the opportunity to read the excellent macroeconomic research from the MacroStrategy Partnership.
As it can be seen below, the exergy economic model puts the economy into this wider context of energy flows. Chart source can be found here.
The key idea is that all interactions of matter involve energy flows. “This is true whether they have to do with earthquakes, the movement of the planets, or the various biological and industrial processes at work anywhere in the world.”
Instead of looking at economic growth as a function of how productive the capital and labour employed are, the exergy economic model explains GDP as a function of labour, capital and exergy. In other words, productivity is equivalent with the energy’s capacity to be converted into physical work. As Niels Jensen writes in his book – “Think of exergy as productivity.”
The reason you can think of exergy as productivity is because they are both measures of the same concept: cost-effective work; albeit from slightly different angles. Building on Ray Dalio’s definition of productivity, take the following example.
Suppose that the workers from country A are equally as educated and produce the same amount of output per hour as the workers from country B. If the workers from country A cost less than the workers from country B, then the country A is more competitive (i.e. it is more productive relative to country B). This is the same thing as saying that if workers of country A use less resources, both financial (money or credit) and physical (energy) than workers from country B, given that the output is equal in both cases, employing workers from country A will provide you with more value above the cost of employing them.
From an exergy economics perspective this can be stated as: the capacity of energy to produce physical work through the thoughts and actions of workers from country A is more cost-efficient because their work uses a lower quantity of energy to produce the same output as the greater quantity of energy utilised in the case of workers from country B. Therefore, between two workers, with the same level of education that produce the same amount of output per hour, more productive (and thus, more competitive) will be the one that costs the economy less resources (not just financial ones but energy too).
This productivity-exergy equivalence also holds true for the process of “producing” energy, particularly from fossil fuels – oil, coal and natural gas remain by far the most important sources of energy worldwide, with 85% of energy supplies coming from fossil fuels in 2017. Therefore, when producing energy is not a productive endeavour (i.e. it is cost inefficient), the production of energy declines, threatening to drag down with it economic development and growth.
Energy production can decline either if demand falls (i.e. consumers find energy prices unaffordable), leading to lower levels of production to meet a lower level of demand, or if supply is cut (i.e. energy producers find it unaffordable to provide the economy with energy or there are political reasons to reduce supply). Finding the equilibrium between these two price levels is what we can call the ‘economic affordability’ of energy.
The role of affordable energy in economic growth
The role of energy has been largely neglected from economic growth models primarily because the models that we are using today have been developed during a time when the world didn’t need to worry about the role of energy too much – until the 1970s oil shocks.
During the Arab-Israeli conflict in the early 1970s, OPEC’s Arab members, led by Saudi Arabia, imposed an oil embargo as retaliation for the support showed to Israel by the US. The embargo involved production cuts and the suspension of exports to the US, as well as other nations that backed Israel. As a result, the price of oil, measured by the spot price of WTI, increased by about 260% above the long-run average in 1974. From 1946 to the end of 1973, the average spot crude oil price was $2.8 per barrel. In 1974 it jumped to $10.1 per barrel, crippling the economy.
This sudden and extreme price increase was a major contributor to the US economic recession during 1973-1975 (the final grey line in the chart below).
We can easily underestimate the important role that affordable energy, tapped into on a massive scale, has played and continues to play in fuelling technological development (i.e. innovation) as well as economic growth (i.e. consumption).
If we look at history, and we do not have to go too far back, we can see that prior to the improvements to the steam engine, which enabled the extraction and transport of oil, the world was advancing much slower than we are today.
An important factor that was holding progress back was the [natural] limit on growth imposed by the sources of energy used: mostly wind (throughout windmills and boats), hydro (through waterwheels and maritime transport) and solar (through food, wood and animal muscle power). These sources of energy were more or less contained by ‘land’ – hence, it is understandable why the classic economists (such as Adam Smith) thought about factors of production as land, capital and labour.
Things changed during the industrial revolution. It was 1859, when the first oil well was drilled in Western Pennsylvania, which was producing 15-25 barrels of oil per day – that was the entire oil production worldwide. In 2017, the world extracted 92.6 million barrels per day. The cost-efficiency of this massive production is paramount for economic development to happen and for economic growth to accelerate (or even to be sustained).
From the perspective of the oil producer, affordability means that it can sell the oil barrel at a price high enough not only to compensate for the cost of discovering the well, extracting, refining and transporting the oil but which also enables future investment in this process as well as in other life endeavours, i.e. the producer must sell oil at a price above the breakeven point, which makes a profit. For example, the average cost of producing one barrel of shale oil in the US is about $60 – if the shale oil price drops below this for a sustained period of time, all things being equal, the producers will have to cut production so the price can reach a new equilibrium at which they make a profit. In reality however, all is never equal – as Niels Jensen shows in “Addicted to Oil?”, the US legal system (Chapter 11) enabled US shale oil producers to restructure and continue production even when the oil price was $30 per barrel between 2014-2015. But not all oil producers around the world benefit from the flexible American legal system.
From the perspective of the consumer, affordability means that the price of oil is low enough that it doesn’t absorb too much of their income, so that they can pay for their mortgage / rent, other utilities, food etc. If the price is too high for the consumer, then demand for oil will go down, lowering the price of the commodity which, in turn, can make oil production uneconomical. This then can result in more cuts to production, meaning that industries won’t have enough energy to sustain the past level of production of goods and/or services. As a result, economic growth comes under pressure.
In my view, unaffordable oil prices, or energy prices in general act as a tightening of resources in the economy – akin to how tighter monetary policy acts for the financial markets. Not high enough prices to justify the cost of energy production means that more resources (labour, finance, technology) are tied up in producing the same (or an increasingly lower) quantity of energy – note that the quality of this energy (i.e. its exergy, or capacity to perform physical work) is also an important factor in determining the productivity of the process of producing that energy. Not low enough prices to enable consumption to pick up (and this includes deferred consumption, i.e. savings) or be maintained means that more income (and / or credit borrowed against future income streams that can also diminish if energy prices remain too high) go towards paying for energy instead of going into savings (pensions and other investments) or other goods / services. Either way, the economy needs affordable energy to develop and grow.
Energy’s cost-efficiency can also be eroded if the energy itself does not do the amount of physical work needed to pay for itself. In other words, as exergy decreases, energy production becomes an unproductive endeavour, eventually rendering it unaffordable. Exergy deteriorates in the process of transforming energy from its raw (or primary) form (e.g. fossil fuel, solar, wind, water) into useful form (e.g. motion, heat, light).
Data on exergy and economic growth and / or development is difficult to come by because this is an area of economics that is not studied enough. However, we can create a rough proxy for energy efficiency by looking at how much GDP we get per energy consumed.
At the first glance it would appear that the world is highly productive in allocating energy resources – for 1 Btoe of energy consumed in 2017 the global GDP produced was c.40% higher than that produced for the same amount of energy consumed in 1980. However, my intuition tells me that there is more to this than meets the eye.
For example, from 1980 to 2017, US total debt (public + private business debt) increased by 20x while US GDP increased only by 7x. In fact, the US GDP per US total debt continued to decrease since 1966. In the chart below, the LH shows the fed funds rate and the RH shows the GDP per Debt.
If the world (overall) would have been allocating energy resources efficiently then would the economy be so dependent on monetary policy remaining accommodative?
The alchemy of credit, productivity and energy
When faced with an unproductive economic process, we have three choices:
a) we invest in R&D / new technology to try and improve the cost-efficiency of that process (i.e. make it better)
b) we do nothing and continue with the unproductive process as it is (in this case resource consumption will eventually deliver a net negative impact to the economy)
c) we do nothing but pretend that we do (which is to keep the process alive through debt – an unproductive process is uneconomic by definition and will eventually stop delivering cash flows; these cash flows can be artificially sustained through “affordable” debt).
Debt in itself is not a bad thing for the economy. In fact, the economy wouldn’t function without debt. As such, we cannot separate the importance of credit cycles from our analysis of productivity and, as we will see shortly, debt has direct consequences on the physical resources that power the economy, not just on the flow of financial capital – this relationship between credit and physical resources is what links debt growth with productivity and thus with exergy.
In A Template for Understanding Big Debt Crises, Ray Dalio outlined the pattern of the archetypical debt cycle – to keep things short, I will not make the difference between an inflationary and a deflationary deleveraging (which happens in the second half of the debt cycle).
Debt cycles start in benign economic times when there is enough slack in the economy and productivity and growth pick up; in this part of the cycle debt doesn’t grow faster than income as it finances productive endeavours that more than pay for themselves (i.e. generate cash flows that can support the credit interest payments and also provide further investment to other parts of the economy).
As the economy continues to do well, incomes grow, the stock market rises and people and businesses become more creditworthy – as such, more credit is extended to them by willing lenders. At some point in the cycle productive allocation of debt starts to turn into an unproductive endeavour – the hallmark of this shift is when credit payments start to grow faster than incomes.
This process can continue for some time as it is kept alive by easy monetary policy (such as low interest rates which makes debt servicing cheap enough). This was also observed by Austrian economist, Knut Wicksell who argued that the danger with credit is that, as the cycle progresses, it will fund more and more enterprises that can only stay alive while credit expands (i.e. zombie businesses).
Eventually however, the unproductive allocation of debt capital stops, usually because monetary policy is tightened in an attempt to fight off inflation; monetary policy tightening increases the cost of debt and adds more and more strain on the cash flows of businesses and households until the economy goes into the deleveraging phase. During this part of the debt cycle, the economy should be cleared of unproductively allocated capital and, if done right, meaning that the policymakers find the appropriate balance between the debt write offs, debt monetisation and redistribution of wealth, the transition towards the recovery phase will be smooth.
However, if the misallocated capital that has been building up during the debt cycle is not cleaned up enough to enable productivity to pick up, the recovery will be a mere mirage. This can happen towards the end of a long-term debt cycle, which usually lasts several decades. Part of the reason is that each time we go through a short-term debt cycle (i.e. a business cycle) we end up with a little bit more debt at the start of the recovery (i.e. at the start of the next business cycle). Eventually, these ‘leftover’ amounts of debt add up to the point that when the ‘final’ deleveraging comes (the end of the long-term debt cycle), orthodox monetary policy is rendered useless – it is interesting to note that Ray Dalio underlined three types of monetary policy through which the economy goes when trying to deleverage – the first is the lowering of interest rates; once interest rates hit zero, policymakers move to the second type of monetary policy which is the monetisation of debt (quantitative easing); and finally, once the impact of quantitative easing on stimulating economic activity diminishes, policymakers move to what it is known as ‘helicopter money’, which is essentially quantitative easing for Main Street.
It is with this part of the long-term debt cycle that we are concerned today, as we find ourselves close towards its reversal. Interest rates remain very low compared to history across many developed nations and quantitative easing has had a more pronounced impact on boosting financial assets rather than productive economic activity. At this point, credit creation is highly likely to be misallocated.
As I wrote in The Wealth Illusion, there are at least two ways to assess whether or not credit has been allocated productively – the growth of debt relative to GDP and the Wicksell spread. Both measures continue to suggest that debt is being misallocated.
The link between debt creation, productivity and exergy has been illustrated by the excellent work done by the MacroStrategy Partnership. Succinctly, once credit starts to be misallocated, if this process is kept alive by easy monetary policy long enough, then, eventually, the economy will reach a point at which any credit creation depletes physical resources – in essence borrowing output from the future.
What this does is lowering productivity, both from the perspective explained by Ray Dalio (see above) and from the point of view of the exergy model. From the perspective of the creditor, debt is a claim on future cash flows.
Here is another interesting way to view debt, from the perspective of the debtor: “A debt is a short cash position—i.e., a commitment to deliver cash that one doesn’t have. Because the dollar is the world’s reserve currency, and because of the dollar surplus recycling that has taken place over the past few years…lots of dollar denominated debt has been built up around the world. So, as dollar liquidity has become tight, there has been a dollar squeeze. This squeeze…is hitting dollar-indebted emerging markets (particularly those of commodity exporters) and is supporting the dollar. When this short squeeze ends, which will happen when either the debtors default or get the liquidity to prevent their default, the US dollar will decline. Until then, we expect to remain long the USD against the euro and emerging market currencies.” Source: A Template for Understanding Big Debt Crises, Ray Dalio, 2018.
If the debt is used productively, i.e. in a way in which the value of what you get (goods/services) is higher than the cost of producing it (cost of capital and of transforming energy from its primary form to its useful form), then future cash flows are safe.
If that value drops below the cost of production, then, all else being equal, future cash flows will diminish. However, they can be artificially sustained through more borrowing (remember, we are talking about an environment where momentary policy is accommodative, i.e. borrowing is cheap). These artificially sustained cash flows mean that credit creation enables physical resources (such as land, machinery, labour etc.) to be tied in unproductive processes of production.
Viewed from the angle of the exergy model, this is equivalent with saying that the energy used by these processes sees a diminished return on its ability to do physical work (i.e. its exergy declines). Human capital, machinery and natural resources employed in these processes, in reality, produce a diminished (and eventually negative) return, which is masked by the credit subsidised cash flows. All of this has been made possible by monetary policy being kept accommodative for too long – too long means that interests rates are low and central banks keep buying assets (or keep reinvesting the proceeds from these assets, i.e. maintain their inflated balance sheet) after economic activity has recovered as demonstrated by various indicators such as bank lending revival, healthy employment indicators and business confidence levels.
The ultimate result of this process, if it is not reversed by somehow reducing the debt burden (which when it grows too big it doesn’t happen without a lot of economic pain and potentially social unrest) is negative growth in the future.
We are now at a point in which we need to consider both the credit cycle and the energy conversion cycle – there are arguments which suggest that the world has already reached peak oil. The challenge with these types of analysis is that the data is subjected to the uncertainty of undiscovered oil resources. However, what is certain is the finite nature of fossil fuels.
The natural limit to growth
Unproductive allocation of resources cannot be sustained. Eventually, either the burden of debt or the depletion of fossil fuels will put the brakes on economic development and growth. There is analysis that suggests that this is already happening.
The chart above shows that the world’s economic growth is returning to its historic trend. Since 1970 both debt and energy consumption have been increasing across the world, while economic output per capita has been declining. There are other factors at play here, which are beyond the scope of this paper. One of them are the values that underpin our society (e.g. how much we value getting a new pair of Gucci shoes for £600 over saving that money).
Renewables could play a more prominent role in our economy but, for now, they are not as efficient as fossil fuels to support the level of production that we have become accustomed to. So we either invest more in R&D to look for alternative sources of energy, or at least to improve the technologies that capture, transform and transmit renewable energy or we cut our consumption.
What prompted me to write this paper?
Two factors contributed to my decision to write this paper. The first one was Ray Dalio’s incredible work on the debt cycle and productivity. Ray’s detailed analysis of what drives economic growth convinced me that the world works in a cause-effect like fashion, akin to a machine – this means that if we understand how the building blocks of this machine interact and what brings these building blocks about, we can form a much more complete picture of how this machine is likely to perform in the future, what may be in need of repairing and what may need upgrading.
The second factor was the work of Gail Tverberg, who put the concept of exergy into the broader context for me and made me realise the importance of finding alternative sources of energy that can enable the world to become less dependent on fossil fuels (due to both their finite nature and their polluting nature).