Mainstream economics seems to have learned little and changed nothing in the last decade, despite the fact that the financial crisis and its aftermath laid bare a number of important issues with its theories and models. Failure to address these issues is making the economics discipline increasingly incapable of informing us about the trajectory and situation of our world.
After a long period of relentless rise, global financial markets seem to have suddenly entered volatile territory. A brutal selloff in global stocks started in early February, which erased all of the prior gains of 2018 and wiped out trillions of dollars of ‘value’ in a matter of days. The selloff was most spectacular in the U.S., with Wall Street experiencing one of its worst weekly tumbles since the 2008 financial crisis – quickly followed, however, by a sharp rebound. Financial pundits the world over are now busy discussing whether this new episode of market volatility is already over or is likely to last, and if it might be announcing a ‘correction’ (a drop of 10% or more from a peak in market indexes), a ‘bear market’ (a drop of 20% or more), or even a full-blown crash. The truth is that no one knows for sure at this stage, and any prediction of how the next few weeks and months are going to play out in global financial markets can only be guesswork at best.
What is more interesting is to observe how quick economists and policy makers around the world have been to serve yet another round of what has become their standard discourse whenever financial markets get suddenly restless: no worries, folks, ‘the fundamentals of the economy are strong’… Of course, one couldn’t really expect them to say anything else right now, since they have spent the last few months explaining that the long economic slump that followed the 2008 financial crisis was finally over, and that the ‘recovery’ was now poised to accelerate. As Harvard University Professor Kenneth Rogoff put it just a few weeks ago, the global economy is well engaged in a process of ‘reversion to mean’, and “we are going to get above average productivity growth and rising investments for several years as the economy normalizes”. Hence, “there is no reason to suppose that the odds [of a recession] are greater than 15% today. On the contrary: There is a very good chance that growth outperforms most of the time the next few years”. ‘The fundamentals are strong’, you know…
The same words were used, if you remember, in 2007 and even during much of 2008, when the world was rushing towards the worst financial crisis in 80 years, which would trigger the worst global recession since World War II. Back then, most economists and policy makers were slow to grasp the causes, magnitude and consequences of what was happening, and many remained stubbornly confident about the economic outlook until the global financial system got to the brink of a complete meltdown. Even then, some contended that the subsequent downturn would likely be mild and short, and surely followed by a speedy return to long-term growth trends. ‘The fundamentals are strong’, you know…
Still a dismal science
Economists, if history is any guide, have a notoriously dismal track record in terms of foresight: not only did the vast majority of them fail to ‘predict’ the Great Financial Crisis and the Great Recession, they also failed to foresee most of the previous financial crises and recessions or even to recognise them until after they had begun. To be fair, it is actually very difficult to understand an organism as complex as the economy and to forecast its evolutions and cycles. The economy is indeed a dynamic, complex system, comprised of a multitude of elements that evolve constantly and interact in various ways, meaning that some of its aspects are likely to elude any interpretation that may be made at any given point in time. Yet the economists’ answer to this difficulty has largely consisted in evacuating whole aspects of the complex economic system from their view and in reducing their field of inquiry to a limitative set of relatively basic components that they contend can be ‘scientifically’ described and measured.
The desire to bring the rigor and objectivity of ‘hard sciences’ into the ‘soft’ field of economics has underpinned the evolution of the discipline since the end of the 19th century, in particular the development of increasingly sophisticated mathematical modelling, of econometrics and financial econometrics. This evolution has made (macro)economics a ‘reductionist’ discipline, which ‘rules’ are not only based on questionable assumptions and simplifications, but also – and more fundamentally – only make sense insofar as whole sets of elements that are constitutive of the ‘economic system’ (i.e. the system by which resources get allocated and goods and services get produced and distributed) remain ignored. A discipline, as a consequence, which ‘progress’ has made it and is still making it increasingly incapable of grasping and accounting for the reality it pretends to study.
This kind of criticism of economics – and of economists – is obviously not new. In fact, economics has never ceased to be called ‘dismal’ ever since it started claiming to be a ‘science’ of sorts in the second half of the 19th century. Of course, economists typically tend to dismiss it as unfounded and uninformed babble by unsophisticated people who fail to grasp the ‘elegance’ of their mathematicised thinking and models. Yet there is mounting historical evidence that these theories and models are largely detached from reality, and the last few years in particular have shed new light on this widening disconnect. Not only because most economists failed to see the financial and economic crisis coming, but because the years since then have failed to reassure about their understanding of what was and is still going on.
One could have hoped that economists would have at least learned something in the last decade, and changed something in their theories and models that could restore trust in their ability to understand reality and where it is headed. Yet it increasingly looks like most of them have learned little and changed nothing. As the global economy now shows signs of recovery, many economists contend that the events and developments of the last decade have not been ‘exceptional’, as Nobel Laureate Paul Krugman recently stated, in the sense that they have not been “clearly inconsistent with widely held views and sustained enough that they couldn’t be written off as aberrations”. The financial crisis has merely been, in the words of Kenneth Rogoff, “a garden variety of a systemic financial crisis”, technically “very normal compared to other deep systemic financial crises”, and which triggered an also ‘very normal’ prolonged economic slump. In other words, economists haven’t learned much from this episode that they didn’t already know, and they haven’t had to change much of their views as a result. Hence, the theories and models upon which economists base their analyses and forecasts today haven’t fundamentally changed since 2007-2008.
Dissident voices have of course emerged in the economics profession since then, criticising some of the key theoretical underpinnings of the New Classical–New Keynesian mainstream, in a more or less radical way. A number of heterodox schools of economic thought have gained in popularity and attracted some attention, proposing alternative views and models. However, they have largely remained confined to the fringes of the discipline, and have failed to trigger a substantial paradigm shift in the profession. That may be because these dissident voices and schools have remained scattered, each focusing on a specific set of issues rather than on outlining a new ‘general theory’ to make sense of economic reality in a holistic way. That may be because the long economic slump has not morphed into a full-blown depression, and therefore hasn’t left mainstream economists with no other choice than to revisit their views. That may be, also, because economics is already too far advanced on its current path to turn back, reassess some of its basic theoretical underpinnings and make changes on the scale that would be required to somehow reconnect the discipline with the real world.
Whatever the reason, the issues with conventional economic doctrine that were laid bare by the crisis and its aftermath remain largely ignored by mainstream macroeconomists. These issues are many, varied and complex, and they have been widely exposed and discussed outside of mainstream economic circles in the last decade. The most fundamental ones, which need to be understood to make sense of where the world stands in 2018, probably consist in a significant flaw (i.e. something that economists get plainly wrong), a major blind spot (i.e. something that economists do not see, or only partially), and a monumental omission (i.e. something immensely significant that economists ignore entirely).
A flaw in the model, so to speak
The most fundamental flaw in mainstream economic theory that was made plain by the crisis probably concerns the behaviour of economic agents and its determinants. A key theoretical underpinning of modern macroeconomics is that economic agents act rationally and in their own self-interest, i.e. that they make choices that are guided by rationality, interpreted as the desire to ‘maximise’ their individual utility, and that their expectations are also rational, based on available information and past experiences. The rational choice theory and the rational expectations hypothesis underpin the frameworks that most economists use for understanding and modelling the economy and markets, as well as the belief of many of them in the existence of a natural economic ‘equilibrium’ or in the natural efficiency and self-correcting virtues of free markets.
Yet the Great Financial Crisis showed in a dramatic way that the expectations and behaviours of market participants and economic agents could drift far away from any kind of assumed rationality. It suddenly appeared, then, that the development of increasingly complex financial instruments and the progressive liberalisation of financial regulation in the U.S. and across the Western world had led to excessive, irresponsible and ‘irrational’ risk-taking across the financial system, and in particular to the reckless piling up of risky assets on bank balance sheets. The market had manifestly not been as efficient as it was meant to be, presumably because the expectations and behaviours of many agents had not been as rational as they were supposed to be. Those who had put their faith in the rationality of self-interested market participants were thus left “in a state of shocked disbelief”, as famously declared by former Chairman of the U.S. Federal Reserve Alan Greenspan to Congress in October 2008. “Yes, I found a flaw”, Mr Greenspan had to admit, a “flaw in the model that I perceived is the critical functioning structure that defines how the world works, so to speak”…
The financial crisis therefore triggered renewed interest in ‘behavioural economics’, which addresses the limits of rationality in economic decision-making and aims to provide a broader and more realistic appreciation of how market decisions are made and of the mechanisms that drive economic choices. Behavioural economics studies the effects of psychological, social, cognitive, and emotional factors on the economic decisions of individuals and institutions, and their consequences for market prices, returns, and resource allocation. Unlike the rational expectations hypothesis, which is a model-consistent hypothesis but has no solid empirical foundations, behavioural economics has a well-documented and growing empirical base.
In recent years behavioural economics has gained in popularity and recognition, thanks in particular to influential works by Daniel Kahneman[i], who was awarded the Nobel Memorial Prize in Economic Sciences in 2002, or by Richard Thaler[ii], who won the prize in 2017. It has gained some influence in finance and also in public policy, as shown by the establishment by several governments of dedicated teams of behavioural scientists tasked with providing insight into the design and implementation of their policies. However, it has yet to really make its way into mainstream macroeconomic thinking and models, which largely continue to assume the existence of perfectly rational agents.
A blind spot for money, debt, finance…
Besides this manifest flaw, the financial crisis also revealed that macroeconomics suffers from a fundamental blind spot for the role of money, debt and finance in the economic system.
Conventional economic theory purports that money is economically neutral in the long run, meaning that changes in the money supply do not fundamentally affect the ‘real’ economy (i.e. the patterns of production, consumption, employment or trade), but only ‘nominal’ variables such as prices, wages, or exchange rates. According to this view, which is still held by neoclassical as well as many Keynesian economists, a change in the quantity of money may generate short-term disruptions due to the ‘money illusion’, but over time the economy then tends to settle back to the same ‘natural’ ‘equilibrium’ and to the same long-term trajectory as before, which are determined by the patterns and trends of production and exchange and are independent from monetary variables. Economists therefore treat money, as Joseph Schumpeter once said, as just “a technical device that has been adopted in order to facilitate transactions”, which “so long as it functions normally (…) does not affect the economic process”.
As the financial crisis and Great Recession made plain, this view is inconsistent with reality: the quantity of money present in the system does influence the long-term patterns and trends of the real economy. It did so before the crisis, as increasingly loose monetary policy and ‘financial conditions’ contributed to the relentless growth of global financial capital and to the inflation of asset price bubbles that distorted long-term resource allocation and created the conditions for the 2008 blow-up; it does so after the crisis, as even looser policy – in the form of interest rate suppression and the flooding the global financial system with monetary liquidity – has been needed to avert a descent into outright depression and then to avoid a repeat of the recession. A looser policy that has had the long-term effect of ushering an era of superabundant financial capital, re-inflating asset prices, maintaining a ‘zombie economy’ on life support and stocking up long-term underlying problems for later – or even, according to many, making them worse.
If conventional economics misrepresents the role of money in the modern economic system, it also remains surprisingly confused about the mechanisms through which money is created, and in particular about the role of credit creation by commercial banks. Most macroeconomists see banks as mere intermediaries between savers and borrowers, and credit as a simple technical mechanism to move money from where there is an excess of savings to where credit is needed. The credit creation mechanism, in their views, supports the macroeconomic identity between saving and investment, but it does not affect aggregate demand and the dynamics of the economy. Bank lending only redistributes spending power from savers to investors, and the rate of credit creation does not fundamentally impact the economy’s performance.
Yet, contrary to this widely held view commercial banks do not simply act, when making loans, as intermediaries lending out to some people the money deposited with them by others, keeping only a fraction in reserve as a precaution. They actually ‘create’ new money by simultaneously writing in their books an asset (the loan) and a corresponding liability in the form of a matching deposit in the borrower’s bank account. Therefore, commercial banks create most of their own deposits themselves by extending credit to borrowers/depositors. When creating new loans and deposits they don’t just ‘multiply up’ central bank money, they are free to provide financing as they decide based on the profitable lending opportunities available to them in a competitive market and on their own appreciation of the related risks. There are of course regulatory constraints on credit creation in the form of prudential requirements and ratios (e.g. capital or liquidity ratios), but within these limits commercial banks can and do create money ‘out of thin air’ by just loaning it into existence, in a transaction that involves no intermediation.
Therefore, and contrary to beliefs that remain widespread even among economists, the vast majority of ‘money’ in the modern economy is not created by governments or central banks but by commercial banks in the form of bank deposits, through the mechanism of lending. This money creation mechanism has fundamental consequences for the economic system. The first is that most money in a modern economy is created as debt, or in other words that most modern money is, in essence, debt-based and hence extinguishable (i.e. it ceases to exist when the debt is repaid). Expanding the money supply therefore means and requires expanding the net amount of debt in the economy (i.e. creating more new debt than what debt is repaid). Inversely, a reduction of the net level of debt in the economy effectively reduces bank deposits and contracts the money supply. A second consequence is that the main purveyors of debt-based money, that is commercial banks, have a significant influence over the shape and performance of the economy, through both the quantity and quality of their lending. When they increase their lending to households and businesses, the amount of credit goes up and there is more money available in the economy. This allows for increased consumption and investment, which in turn creates jobs and expands income and profits and also pushes up the price of assets (e.g. stocks, bonds, houses, etc.), giving borrowers more wealth against which they can borrow still more. In other words, credit growth boosts economic growth, in a sort of upward spiral of prosperity.
The relation between credit growth and output growth is not 1:1, of course, because other factors also influence economic growth, and because lending quality also matters: credit creation can in theory have a positive multiplier effect on output if it targets the comparatively most productive purposes, or on the contrary fail to trigger much growth if it doesn’t. However, all other things being equal, the causality is inescapable in the short term: more credit means more growth – at least as long as the spiral of prosperity goes up; less credit means less growth. If credit growth slows down because banks and/or borrowers decide or have to be more cautious, economic growth will tend to slow down as well. If the aggregate level of credit then stops growing altogether or even contracts because banks and/or borrowers need to reduce their risk exposure, the economy will also tend to contract.
The spiral of prosperity can then even go into full downward mode. Indeed, a net repayment of debts (also called ‘debt deleveraging’) effectively reduces the amount of money in circulation and hence the overall number of monetary transactions. This may push price levels down and depreciate the market value of assets relative to the cost of the debt or borrowing assumed to acquire them. The real value of remaining debts then goes up, which hampers the capacity of economic agents to repay and service them while at the same time increasing the pressure to deleverage. This may turn into a self-feeding deflationary process that some economists call a ‘debt deflation’, i.e. a situation in which massive deleveraging leads to a deflationary downward spiral and widespread financial distress.
In the decades prior to the financial crisis, decreasing interest rates and financial deregulation led to a progressive relaxing of credit conditions, which fuelled accelerating credit growth across the globe, and particularly in Western economies. A three decade-long global debt build-up ensued, which has been dubbed a ‘debt supercycle’ by some economists. The piling up of private debt (i.e. debt contracted by households and businesses) during that period was unprecedented in scale – in fact, it was the largest peacetime accumulation of debt in the world’s history.
This debt binge had two fundamental consequences for the global economy. The first is that it contributed to fuel and accelerate output growth beyond what would otherwise have been possible, thanks to a ‘leverage’ effect – i.e. the use of borrowed money, as opposed to equity, to finance operations or investments, with a view to increase potential returns. The second is that it ushered a wide-ranging process of ‘financialisation’ of the economy in the decades prior to the 2007-2008 financial crisis. Financialisation can be defined as the process by which financial motives, financial markets, financial actors and financial institutions take an increasing role in the operation of an economy. It generates a pattern of accumulation in which profit making occurs increasingly through financial channels rather than through the production and trade of goods and services. It typically leads to an increase in the size and importance of the financial sector relative to the overall economy, which profoundly influences both corporate behaviour and economic policy. Eventually, it ends up making some financial institutions ‘too big to fail’, as any such failure would have massive and devastating rippling effects across the financial system and the wider economy.
The ‘fuel’ of the financialisation process is debt creation, or rather the mass commodification of debt, i.e. the turning, through securitisation mechanisms, of debt contracts and debt relations into commodities that are bought and sold for a profit by financial investors. Financialisation is only possible, in fact, if low-cost credit and leverage massively expand, and if their effect gets multiplied and their risks spread through mass securitisation, which is what occurred in the decades leading to the Great Financial Crisis. However, just like pretty much everything in human affairs the expansion of debt and its mass commodification through securitisation are subject to the law of diminishing returns, meaning that adding more debt – and pushing it around in the form of increasingly sophisticated asset-backed (i.e. debt-based) securities – inevitably tends at some point to yield lower incremental per-unit returns. Or, in other words, an ever-growing amount of debt and an ever-growing use of securitisation are required to obtain a same level of returns.
This is what happened during the ‘debt supercycle’ that preceded the financial crisis, when the growth of credit quickly outstripped the growth of economic output, signalling a decline of marginal ‘debt productivity’ (i.e. the amount of output growth obtained from each new unit of debt, or in other words the ratio of economic growth to debt growth). In fact debt productivity has been trending down for decades, meaning either that financial institutions have increasingly been failing to allocate credit to its most productive uses, that potential productive uses have been rarefying overall, or a mix of both. Whatever the reason, declining debt productivity led to a sharp acceleration of the debt build-up at the turn of the 21st century, which as financial actors kept looking for ever-higher performance and returns contributed to sending the securitisation frenzy into overdrive, fuelling speculative bubbles and spreading increasing risks across the financial system. The 2007-2008 financial crisis erupted when the U.S. housing bubble burst, triggering cascading ‘credit events’ that led to a seizure of global credit markets. Banks suddenly reduced their lending, first to each other and then to the rest of the economy, causing a credit crunch that quickly spread across the world and threatened to bring down the massive global web of asset-backed/debt-based securities after the downfall of ‘too big to fail’ investment bank Lehman Brothers.
To prevent the ensuing deleveraging process from spiralling into an uncontrollable debt deflation, the U.S. Federal Reserve – soon followed by other central banks – slashed policy (i.e. short-term) interest rates to almost zero and embarked on the large-scale unconventional monetary experiment of ‘Quantitative Easing’ (QE), i.e. the purchase from the market of massive amounts of government or corporate securities using newly created central bank money. The objective of Zero Interest Rate policy (ZIRP) and QE was to prevent the price of financial assets and securities from spiralling down and to support them over time, with a view to lower long-term borrowing costs and ultimately stimulate credit-based investment and spending. In fact, QE quickly ‘reflated’ financial assets and then propped them up during several years, which probably helped banks and other financial institutions to repair their balance sheets faster and more easily than would otherwise have been possible. This, in turn probably contributed to contain the credit crunch and helped credit flows to progressively resume across the economy. QE thus helped to prevent a bank balance sheet clean up from morphing into a full-blown debt deflation – and hence the ‘Great Recession’ from turning into a new ‘Great Depression’.
However, by preventing the pre-crisis stock of debt from deflating, QE also prevented the deleveraging process that typically follows a financial crisis from running its course. It made it possible for the overall indebtedness of non-financial economic agents in advanced economies to remain close to the high levels reached in the run-up to the crisis or even to rise slightly in some countries, leaving little room overall for further rapid expansion. Hence, the growth of private debt slowed down and never returned to pre-crisis levels in advanced economies, because lending standards to households and small businesses became more stringent after the crisis, but also and mostly because the level of indebtedness of economic agents was already excessive for them to re-start accumulating debt at the same rate again. The lack of significant deleveraging led to the persistence of a massive ‘debt overhang’ that probably acted as a major constraint on economic growth in the years since the crisis. According to some ‘unorthodox’ economists like Pr. Steve Keen, the resulting ‘drought’ in credit expansion has even been the main cause of the persisting dearth of economic growth that followed the recession: too little new debt was being created because too much old debt was still on the books…
As surprising as it may seem, all these developments largely elude conventional macroeconomic theories and models. Most macroeconomists have traditionally paid little attention to the role of credit in driving economic growth, and to the role of commercial banks in driving credit growth. The reason for this is that they still fundamentally see banks as mere intermediaries between savers and borrowers, and credit as a simple technical mechanism to move money between those two groups. In their views, the fact that a debt on someone’s balance sheet necessarily appears as an asset on someone else’s balance sheet means that, on aggregate, “debt is money we owe to ourselves”, as Paul Krugman put it, and therefore the level of debt as such does not fundamentally affect aggregate demand and economic growth – only its distribution matters.
As a consequence, banks and the financial sector do not even appear in most mainstream macroeconomic models. These models typically contain a small number of ‘representative agents’, such as a household, a non-financial business and the government, but no financial sector, which remains ‘exogenous’ to the economic system. When banks exist in the models they usually appear as simple intermediaries transferring funds between savers and borrowers in a general equilibrium setting. This does not correspond to the reality of modern banking, in which banks are profit-seeking firms that make loans opportunistically and can independently create new credit and allocate it in the economy, thus significantly influencing its shape and trajectory. The amount of credit extended by banks, or rather the net change in the level of debt (net credit creation) is, in fact, a crucial variable affecting the dynamics of the economy. And, in a ‘financialised’ economy, the financial sector plays a crucial role in influencing the shape, performance and stability (or lack thereof) – of the economic system – something that remains poorly understood in mainstream macroeconomics.
The Great Omission
As important as this blind spot for money, debt and finance might be, it is not even the most fundamental issue with macroeconomics. Something deeper and more significant lies beneath the discipline’s increasing inability to account for reality, something that economists flatly ignore.
In the years that followed the crisis, the economic debate came to focus on the weakness of the recovery and the difficulties in re-starting the growth engine. Besides the persistence of a debt overhang and the resulting lack of rapid credit growth, other explanations given have included the long-lasting effects of the crisis itself. Economic history indeed suggests that deep recessions, and recessions caused by financial crises in particular, can have persistent negative impact on growth via their negative effects on capital accumulation, on innovation and competition dynamics, as well as on labour input (i.e. unemployment and under-employment) and productivity (e.g. skills and knowledge deterioration). Some economists use the term ‘hysteresis’ – borrowed from physics – to designate these long-term effects of deep recessions, which leave persistent scars on the economic tissue and tend to shift the economy down to a lower expansion path.
However, a number of economists then started to recognise that the ‘Great Malaise’ that followed the Great Recession probably reflected something deeper and more fundamental than just the effects of the crisis. Something that predated the crisis and actually contributed to its occurrence. That something is the long-term slowdown of global economic growth, which started long before 2008.
Output growth has indeed been trending down for decades. World economic growth in terms of real GDP (i.e. the measured value of economic output adjusted for price changes – inflation or deflation) has been slowing down for several decades, averaging 3% in the decade 1991-2000 vs. 5.5% in the decade 1961-1970, 3.9% in the decade 1971-1980 and 3.2% in the decade 1981-1990 (source: www.worldeconomics.com). A rebound to 3.7% took place in the decade 2001-2010, which was almost entirely attributable to the catch-up growth in emerging economies – principally China – that resulted from the globalisation process. But in advanced economies the slowdown continued and accelerated, on aggregate as well as on a per capita basis. Just as people and businesses in the West were piling up debt increasingly irrationally and the financial sector was going rogue in many countries, the ‘real economy’ – i.e. the part of the economy that is concerned with actually producing goods and services, as opposed to the part of the economy that is concerned with buying and selling financial assets – was slowly but steadily decelerating.
This ‘Great Deceleration’ occurred despite the fact that, for a time, the increasingly massive recourse to debt financing of investment and consumption boosted economic growth beyond what would otherwise have been possible. And it started long before indebtedness levels reached the heights where they would themselves start acting as a drag on growth. It however contributed to making the debt binge increasingly unproductive and unsustainable, despite a continuous decrease of interest rates, as the disconnect between the mountain of obligations that was getting relentlessly created in the financial sphere and the underlying wealth creation occurring in the real economy was growing wider and wider. Long before 2008, the world was getting set up for an inevitable financial blow up…
The ‘Great Deceleration’ also occurred despite the fact that what many economists – those of the Classical type – see as major impediments to economic growth (i.e. obstacles to free enterprise, free markets and free trade) have been significantly and consistently reduced over the last decades. Economic policy has in fact tended to be geared towards privatisation, deregulation, and lower taxes in most Western countries since the 1980s, and the globalisation process has tended to reduce government size and to enhance economic freedom across developed as well as emerging economies, making it possible to dramatically increase the global flow of goods, services, and capital. Globally, government-imposed barriers to entrepreneurship and business, as well as to trade between nations, are probably lower today that at any time in human history.
What, then, caused the growth slowdown? For some economists – mostly those of the Keynesian type – it primarily results from a long-term decline of demand in the economy, i.e. lower consumption spending and lower investment spending in physical and human capital, leading to a lower quantity of goods and services demanded in the aggregate. A number of factors have been identified for explaining this demand decline, starting with population ageing and the sharp rise of income and wealth inequality in many Western countries[iii], which tends to reduce labour force productivity and to hamper people’s participation in the economy as earners or consumers, while concentrating income and wealth in the hands of rich households with lower propensities to consume. The resulting decline of aggregate demand has led some economists to talk of a ‘Japanification’ of the West, or, more recently, of a descent into ‘secular stagnation’, i.e. a prolonged period of slow growth induced by structural demand deficiency.
According to former U.S. Treasury Secretary Lawrence (Larry) Summers, who revived the secular stagnation hypothesis in 2013, the chronic demand shortfall originated long before 2007-2008, and resulted from the combination of three long-term developments: the build-up of a ‘savings glut’ (due to high levels of saving by pre-retirement baby-boomers, rising income and wealth concentration, the hoarding of cash by large corporations running financial surpluses, and the increasing hoarding of reserves by governments and central banks in several parts of the world); a dry up of investment opportunities (due to a sharp decrease in the price of capital goods, continuous technological disruption, and slower labour force growth); and persistently low inflation. According to Mr Summers this combination progressively depressed aggregate demand and constrained growth in advanced economies. It also brought the ‘natural’ or ‘neutral’ real interest rate – i.e. the real (inflation-adjusted) interest rate that supposedly balances saving and investment at full employment – ever lower, to the point of becoming negative in some cases, making it increasingly difficult for an expansionary monetary policy to translate into fluctuations in price levels or to stimulate economic growth. In fancy macroeconomics speak, at the ‘zero lower bound’ injections of money fall into a ‘liquidity trap’ and exacerbate the chronic excess of saving over investment, which tends to further bring down the natural interest rate and the ‘non-accelerating inflation rate of unemployment’ (NAIRU)… Loose monetary policy however makes access to credit ever easier, which drives investors to take greater risks in search of yield and thus raises asset prices, leading to financial instability and often to the inflation of asset bubbles, which eventually burst. The result, for Mr Summers, is an economy whose normal condition is one of inadequate demand and that unless counteracting measures are adopted can only get anywhere close to full employment when it is being propped up through asset bubbles…
Besides this ‘demand side secular stagnation’ hypothesis, a supply-side view also emerged in recent years, which contends that the ‘Great Deceleration’ primarily results from the slowing pace of technological change and innovation. According to some economists, advanced economies have indeed already exhausted the ‘low-hanging fruits’ of economic growth[iv], which started vanishing during the last forty years as innovation was slowing down and a technological plateau was being reached. This view was famously promoted by American economist Robert Gordon, who argued in a widely discussed book[v] that growth was on a long-term downward trend mostly because of the diminishing returns of innovation. The so-called ‘Third Industrial Revolution’ based on information technology (IT), Gordon said, is fuelling a great wave of technological innovations but is failing to boost productivity on the same scale as great innovations from the past, such as electricity or the internal combustion engine. The last big spurt in productivity, which started in the mid-1990s and was driven by the growth of the Internet, has fizzled since the early 2000s. The growth slowdown, in Gordon’s view, is likely to persist and get worse, which could conceivably lead to the overall disappearance of economic growth after 2050 or 2100. The rapid economic expansion experienced over the past 250 years, he argues, might turn out to be a unique episode in human history – or in other words, economic growth might in fact be a very exceptional situation, rather than the ‘normal’ or ‘natural’ state of affairs that economists, policy makers and most people in the West commonly assume.
Even if they do not necessarily adhere to Mr Summers’ gloomy assessment or to Mr Gordon’s sombre prognostications, many economists and policy makers seem to have accepted that, for a number of supply-side and demand-side reasons, ‘potential output’ growth – i.e. how rapidly the production of goods and services can expand without increasing inflation – has been trending down for decades in advanced economies, and even at global level since the crisis. Yet when analysing the causes of the phenomenon, they tend to stick to what textbook economics tells them determines an economy’s productive capacity: the supply of two factors of production – labour and capital – and how productively they are used. The productive capacity expands when either the supply of these factors or the productivity of their use grows. When either supply of productivity fail to grow, then a downward pressure is exerted on output growth.
Modern macroeconomists can in fact debate endlessly about whether output supply creates demand or the other way around – which is probably little more than a chicken and egg discussion – and about the set of consequences that either one or the other proposition entails, yet they tend to agree on the fact that in any case the production of economic output and its evolution are essentially a ‘function’ of capital and labour. Economic growth, in their view, essentially results from the expansion of the supply of labour and capital inputs and from the rise of the productivity of their use. Labour and capital are the only two factors of production that determine and constrain the patterns of both productivity and scarcity. Most other inputs into the economic process are considered as ‘exogenous’, secondary and largely immaterial to the long-term trends of the economy.
There is a pretty significant issue with this view, though: it makes a rather poor job of accounting for the historical reality of economic growth, which has in fact been much larger since the dawn of the Industrial Revolution than whatever can be modelled with just labour and capital. The economists’ response to this issue has consisted in attributing this unexplained change in output growth after accounting for the effect of capital and labour to something called ‘total factor productivity’ (TFP), which is supposed to represent the economy’s long-term technological progress. This TFP cannot be observed but can only be estimated as corresponding to the large gap between real economic growth and the output growth estimated from capital and labour improvements alone. This gap has come to be known as the ‘Solow residual’, after American economist Robert Solow who first theorised it in the 1950s. Yet there are two significant problems with this theory: the first is that the Solow residual is hardly ‘residual’ at all, since it historically constitutes by far the larger part of economic growth, and thus attributing it to something that cannot be observed is only, as American economist Moses Abramovitz once said, a “measure of our ignorance”; the second is that attributing the growth of ‘total factor productivity’ to the effects of technological progress simply does not square with contemporaneous reality.
Indeed, technological progress seems to have been accelerating over the last decades, which have seen spectacular advances in computer-related fields, as well as in biotech, life sciences, robotics, nanotechnologies, neuroscience, etc. In particular, the Internet has become ubiquitous and has vastly enhanced access to information, as well as changed the way we work, trade, shop, communicate, live and even think. Technological progress shows no sign of abating – some tech gurus even argue that a number of new technologies (e.g. quantum computing, artificial intelligence (AI), robotics, additive manufacturing, and synthetic or industrial biology) are now advancing exponentially and yielding ‘accelerating returns’. Yet those supposedly accelerating returns are not appearing in productivity and output measures: the ‘digital revolution’ has so far failed to deliver ‘The Long Boom’ that was expected or promised by some at the turn of the 21st century, output growth in advanced economies has petered out rather than accelerated, and productivity growth has slowed down markedly, which is causing a lot of head scratching among economists.
Some argue that traditional productivity and growth measures might be missing out on a lot of the ‘value’ created in the modern, digital economy, in which the accelerating pace of economic change makes it increasingly difficult to put a price on economic output, and therefore that actual output, productivity and income growth might be underestimated. Others contend that the long-awaited productivity miracle is around the corner, that it’s just a matter of time before the various elements of the technological boom coalesce into a spiral of exploding productivity and economic growth – and of course that we need to step up our investments in innovation to get there faster. If the ‘Third Industrial Revolution’ already fizzled out, then let’s just pretend a fourth one is getting underway and will do the trick…
Yet the apparent ‘paradox’ of a productivity slowdown in the midst of a tech boom might also suggest that, as Robert Gordon and others have claimed, technological progress could actually yield diminishing rather than accelerating economic and productive returns. In other words, in a society that has already reaped the low-hanging fruits of technological innovation, ideas seem to be getting harder and harder to find – or at least the ‘good’ ideas that may bring about real productive and economic benefits, especially in economies that have moved away from manufacturing activities and towards service based activities in which technical changes tend to yield lower productivity gains. If that is the case, the ‘Fourth Industrial Revolution’ could end up being even more of a disappointment than the third one…
What the ‘productivity paradox’ suggests as well, and most importantly, is that technological progress as such may only underpin part of the ‘residual’ share of output growth that cannot be estimated from capital and labour improvements, and therefore that other things than labour, capital and technological change might influence the patterns of ‘total factor productivity’. Or, in other words, that contrarily to what economists believe labour and capital are not the sole defining factors of economic output, but that other factors are equally, or perhaps more, important.
The most fundamental input into the economic process that economists commonly overlook is, by a wide margin, energy. They typically tend to see energy as just a secondary input, “an input like other inputs” as Paul Krugman puts it, which can be substituted by other inputs and which importance to the economic process is limited to its ‘cost share’ (i.e. the level of energy expenditures as a share of GDP, typically very high in non-industrialised economies and that goes down sharply when embarking on a modern development path, then remaining relatively constant over time in developed economies at between 7% and 10%). Yet once again this view doesn’t square with reality. Far from being a secondary input, energy is a fundamental input that is needed for all human activities, without exception. There is not a single human activity, let alone economic activity, which can take place without energy use. In fact, capital and labour are functionally inert without energy input. It is only through an input of energy that they can be brought together to produce and distribute goods and services, and the productivity of their use therefore depends to some extent upon the quantity, quality and productivity of this energy input. As a consequence, energy should probably be considered a fundamental ‘factor of production’ just like labour and capital.
However, mainstream economics largely ignores the central role of energy in the economic process. This ignorance probably comes down, at least partly, to a difficulty to apprehend what energy actually is. Energy is indeed a word with many meanings yet no universal definition, something that is both intuitively obvious yet abstract and complex. It comes from various sources and is used in various forms that are described and measured in different units. Yet all forms of energy are fundamentally dimensions of the same thing: the capacity of a physical system – be it a human or animal body, a car or a plane, a machine or a computer, a factory, a power grid, an economy, etc. – to perform work. Obtained through food calorie intake, this capacity enables human beings, helped by working animals, to perform a number of physical and intellectual activities. Obtained through other sources of energy, it enables various types of machines – and yes, that includes computers and all the devices and systems that underpin the supposedly ‘dematerialised’ digital economy – to perform a wealth of activities that go far beyond the capability of human or animal muscles and brains.
Energy is also what makes it possible for humans to transform matter, and hence to access, convert and use the biological and physical material resources that are also required for goods and services to be produced, distributed and consumed (e.g. plants or plant-based materials, water, minerals, etc.). The ability to access and use those material resources in turn impacts the quantity, quality and cost of accessible and usable energy resources, and therefore a certain degree of interdependence exists between the accessibility and usability of energy and the accessibility and usability of matter in the economic process, which can result in mutual enablements as well as mutual constraints. A close relationship exists, for example, between the energy and metals sectors, as significant quantities of metals are used by the energy sector and significant quantities of energy are used for metal extraction and processing. A strong nexus also exists between energy and water systems.
The economic process can thus, overall, be conceived as the process by which human societies procure, transform and use energy and matter to create, distribute and consume the goods and services that provide sustenance, security, comfort, mobility and entertainment to human beings. Energy – understood as the capacity to perform work and to transform matter – is thus required for any economic activity to take place, meaning that growing the economy typically requires procuring and using increasing quantities of energy. In fact, economic growth has historically been closely correlated with the use of ever-growing amounts of energy. Ever since the dawn of the Industrial Revolution, all countries that have experienced economic growth have increased their energy use at rates similar to the growth rates of their output. A slight disconnection has occurred in the last decades in advanced economies, but which can be largely attributed to the ‘offshoring’ of energy-intensive industries and the concomitant switch to service activities made possible by the globalisation process, accompanied by the increasing recourse to debt-fuelled growth. At global level, the correlation between output and energy growth remains very strong. A growing body of research[vi] even tends to show that, rather than a simple correlation, the relation between energy use and output growth might be one of co-integration (meaning that the two tend to converge over a relatively short period of time) or even of bi-directional causality, and that energy’s contribution to economic growth goes in any case well beyond the ‘cost share’ of primary energy that is commonly considered in conventional economics. Historically, energy use is either the cause or the facilitator of economic growth. When properly integrated into production functions, the production factor energy indeed accounts for most of the historical growth that mainstream economists attribute to ‘technological progress’, and therefore the inconvenient ‘Solow residual’ largely disappears. Energy, in fact, is ‘The Great Omission’ in economic science.
The mechanisms by which energy underpins economic growth are multiple and complex, and not necessarily fully understood yet. To summarise, energy’s contribution seems to be a factor of the availability and affordability of energy, and of the efficiency of its use, but also and probably more fundamentally of the physical properties of available energy sources, of their ‘energetic quality’ and of their ‘energetic productivity’. Energy indeed comes from many sources, including fossil energy (oil, coal and natural gas), nuclear energy, and ‘renewable’ sources (wind, solar, biomass, geothermal and hydropower), which have vastly different properties in terms of energy density, power density, versatility of use, fungibility, storability, transportability, convertibility, or scalability. These different properties give them vastly different capacity to generate economic and social value, especially when problems of scale and resource dependency are taken into account. Energy sources also have vastly different energetic quality (i.e. capacity to be converted into ‘useful work’ that can effectively contribute to the economic process) and energetic productivity (measured as output energy per unit of energy consumed in the extraction/transformation/transport and delivery process). Physical properties, energetic quality and energetic productivity vary widely not only between energy sources, but also across geographies and over time – and not necessarily in linear ways.
A growing body of research[vii] shows that, ever since the Industrial Revolution, productivity and economic growth in industrialised and emerging economies have largely been driven by the increasing availability of cheap and high quality forms of energy inputs, by the rising efficiency of their conversion to useful or productive work, and by the increasing availability of ‘net energy’ or ‘surplus energy’ (i.e. energy available to do other things than finding, extracting, processing, converting, transporting and distributing energy) obtained from highly ‘productive’ energy sources (i.e. fossil fuels, which provide over 80% of the world’s energy use – a figure virtually unchanged over the last three decades). However, studies suggest that in recent decades these trends might have stopped and even started to reverse. In fact, once cheap and abundant forms of energy inputs – in particular oil, the ‘lifeblood’ of the modern global economy – are becoming progressively scarcer, making their extraction and use more complex, more costly and more energy intensive.
This ‘depletion’ phenomenon does not only constrain the production growth of energy inputs and push their procurement costs up, it also induces a degradation of their energetic quality and energetic productivity, as the best quality and most productive resources tend to get used up first. The declining energetic quality of the world’s major energy resources (i.e. fossil fuels) results in a weakening capacity of the energy supply to power productive work, which might be a significant contributor to the ‘mysterious’ slowdown of productivity growth in recent years. Meanwhile, the decline of their energetic productivity exerts a growing constraint on the amount of ‘net energy’ that can get delivered to the economy. A rising share of the total energy input indeed has to be dedicated to finding, harnessing, transforming and conveying energy to meet a growing demand, while the share that can be made available for other uses (i.e. the net to society) tends to decrease, constraining ‘discretionary’ energy investments and consumption, and hence economic prosperity. This relative decrease of the energetic productivity of the world’s major energy resources appears to be only partly offset by technology-enabled efficiency gains and substitutions, and the supply of energy available for doing other things than getting energy is thus becoming increasingly constrained over time, which tends to erode potential economic growth at global level.
Besides these mechanisms, energy also impacts economic growth in other ways. First, the phenomenon of depletion and its consequences do not only affect the world’s most widely used energy resources, it also affects all non-renewable and partly-renewable natural resources from which most material inputs into the economic process are obtained. These resources are stock-based and therefore exhaustible, and as their use increases their procurement tends to become more expensive and resource-intensive while their quality and productivity tends to go down. Just as for energy resources, this process is only partly offset by technology-enabled efficiency gains, tends to weigh on output and productivity growth as it unfolds, and may in some cases end up making the use of some resources uneconomic (i.e. unprofitable). Energy plays a key role in this process as the accessibility and usability of energy and the accessibility and usability of other material resources in the economic process are largely connected and, to some extent, interdependent. A parallel and simultaneous depletion of energy resources and of other natural resources is therefore likely to have compounding negative effects on economic growth.
Second, the use of energy and matter in the economic process always and inevitably generates various types of environmental impacts, which also influence the capacity of the economic system to expand. The extraction, transformation, transport and use of biological and physical natural resources – including energy resources – as well as the production, distribution, consumption and disposal of goods and services that they make possible, generate various types of ‘waste’ or ‘pollution’, which are returned to the biophysical environment. This use of the environment as a ‘sink’ for waste energy and matter is considered as an ‘externality’ in mainstream economics, i.e. an element that is external to the economic system and for which producers and consumers bear no direct costs. Yet this capacity for producers and consumers to ‘externalise’ the environmental costs of their growing economic activity to the ecosystems used as waste ‘sinks’ has in fact constituted a key enabler of economic expansion. Had producers and consumers been unable to externalise these costs and forced to bear them directly, economic growth over the last two centuries would have been significantly constrained.
Historically, the degradation of the biophysical environment resulting from human activity is therefore a feature of the economic growth mechanism, not a bug. The amount of waste and pollution resulting from human activity has thus grown phenomenally over the last 200 years or so, in sync with the relentless rise of energy and matter use and with exponential economic growth. The use of the environment as a ‘sink’ for waste energy and matter, however, tends to negatively impact the biophysical ecosystems from which energy or material resources are obtained, which in turn can reduce society’s capacity to procure or use them. The costs externalised to the environment are significant and growing or compounding over time, and they can end up weighing on the capacity of the economic system to develop and expand, or even to sustain itself. Beyond a certain point, the negative environmental externalities of the economic process (including climate change resulting from the burning of fossil fuels) cannot be ignored anymore, and social pressure grows to ‘internalise’ them, at least partly, into the price system (e.g. through taxation, regulation, or subsidies). This is what happened over the last few decades with the timid development of a system of environmental protection in Western countries, which inevitably imposed some restrictions and/or costs on economic activity and therefore constrained economic expansion (and/or incentivised the offshoring of energy-intensive economic activity to other countries). It should therefore be no surprise that, in the U.S., the repeal of this system of environmental protection is now one of the key priorities of the Trump Administration to ‘Make American Growth Great Again’.
Overall, the world’s economic growth story since the Industrial Revolution is, to a large extent, an energy story: that of the discovery and growing use of fossil fuels, i.e. sources of energy that were far more abundant, powerful, economic, convenient and versatile than anything humankind had been able to use until then. Fossil fuels not only provided human societies with fantastically increased quantities of energy/capacity to perform work at low cost, but also with energy inputs that were of much higher energetic quality and far more energetically productive than previous energy sources. Energy inputs, also, which environmental damages could be conveniently ‘externalised’ to the biosphere and ignored for a long, long time.
Similarly, the global economic and productivity growth slowdown of the last decades is, by and large, an energy story: that of a world running into the diminishing returns of the fossil fuel bonanza. The diminishing returns of fossil fuels’ abundance, power, affordability, convenience and versatility. The diminishing returns of fossil fuels’ energetic quality and productivity. The diminishing returns, also, of our capacity to dump the waste and trash of our fossil-fuelled civilisation into the biosphere. The exponential growth dynamics set in motion by the Industrial Revolution slowed down as a result, signalling the probability of an upcoming peak and possible reversal. Robert Gordon’s intuition that economic growth could conceivably stop altogether during this century might thus turn out to be right, yet for reasons other than those he had in mind.
The growth and productivity slowdown ‘mystery’ is therefore not so mysterious after all, for those who wish to see. It results to a significant extent from biophysical factors that include the depletion of the world’s main sources of energy and matter, which over time tend to raise the acquisition costs, constrain the quantity and degrade the quality and productivity of the flows of energy and natural resources that can be delivered to the economic process. These also include the constantly increasing costs of some of the side effects of the economic process, in particular multiple types of environmental degradation, which increasingly need to be ‘internalised’ into the economic system. These biophysical constraints are progressively eroding the world’s capacity to create additional wealth, and therefore constraining productivity and output growth.
These developments totally elude mainstream economics, which ignores the biophysical foundations of the economic process. It wasn’t always that way, though, as the 18th and 19th century founding fathers of what was then called ‘political economy’ perfectly knew that ‘land’ – as a representation of natural resources – was a fundamental factor of production, just as labour and capital. The evacuation of the natural world from the economic equation came later, in the 20th century, when the bonanza of energy and matter unleashed by the use of fossil fuels blinded economists into believing that energy and natural resources were universally available and affordable, and thus were unimportant to the patterns of productivity and scarcity. Economists, in other words, drew erroneous conclusions from what was fundamentally an exceptional situation. They chose to become ignorant.
Should they miraculously re-acknowledge the biophysical foundations of the economic process, economists would then realise that their discipline in fact ought to be not just a social science but rather a combination of social and natural sciences. That economic activity is to some extent constrained, at global level, by inescapable physical laws that govern how energy can be converted into useful work and transform matter. That economic growth is not a natural state of affairs but the result of very specific conditions regarding the energetic and material underpinnings of human activity. That economic growth and environmental degradation are, historically, two sides of a same coin, and hence that further economic growth is unlikely to be the appropriate answer to the compounding environmental crises that growth has so far been generating. They would realise, also, that the transition away from fossil energy, made inevitable both by the ongoing depletion of fossil fuel reserves and the need to contain climate change resulting from fossil-fuelled human activity, is far and away the biggest economic challenge we face this century, which will profoundly impact the structure and functioning of human economies and societies in multiple ways. They would dedicate their intellect and energy to finding sensible answers to this monumental challenge… Needless to say, this is not happening and will not be happening. Instead, economists will keep scratching their heads about the productivity mystery, arguing about ways to re-start or boost the growth engine through either austerity or government deficits, and building intellectually ‘elegant’ but reality-optional growth models…
A richer picture of the human ‘economy’
To understand and make sense of ‘The World in 2018’, we therefore need a richer picture of the human ‘economy’ – from ancient Greek oikonomia: ‘household management’ – than what modern economics is capable of providing. Economists are in fact too busy arguing among themselves about the right way to look at things to realise that they’re not actually looking at the right things.
To draw this richer picture we need to integrate the fundamental dimensions that are missing from established economic science: the behavioural dimension, the financial dimension, and most importantly the biophysical dimension. We also need to integrate further dimensions that are also inadequately accounted for by economics, such as the ‘global’ dimension (i.e. the patterns of international trade, capital flows, debt relationships, or political power at global level, which have very different impacts on various national and local economies’ structures and workings), and the ‘cultural’ dimension (which constrains the possibility of outlining economic rules having universal value). Most of all, we need to somehow bring these dimensions together in order to develop a more acute consciousness and comprehension of the road we are travelling. This, in fact, is what we will try to do in the next instalment of this series.
To be continued…
[ii] Nudge: Improving Decisions about Health, Wealth, and Happiness, by Richard H. Thaler and Cass R. Sunstein, Yale University Press, April 2008
[iv] The Great Stagnation: How America Ate All the Low-Hanging Fruit of Modern History, Got Sick, and Will (Eventually) Feel Better, by Tyler Cowen, Dutton, January 2011
[v] The Rise and Fall of American Growth: The U.S. Standard of Living since the Civil War, by Robert J. Gordon, Princeton University Press, January 2016
[vi] For an overview, see: Energy and Economic Growth: Why we need a new pathway to prosperity, by Timothy J. Foxon, Routledge, October 2017