The Economics of the Very Long Run: From Fire to Finance in Two Million Years

Tuesday, 5 February 2013 - 6:00pm
Barnard’s Inn Hall


An economist's view of two million years of human history, covering milestones such as fire, trade, private property, agriculture, towns - and writing, accounting and finance.  In particular Professor Schaffer examines why living standards for the average human being changed so little over the first two million years, and so much over the past two hundred years.


Transcript of the lecture

5 February 2013

Economics: From Fire to Finances
In Two Million Years

Professor Mark Schaffer


First of all, I’d like to start with some thank-yous – to Gresham College, for hosting this lecture; to Michael Mainelli, for suggesting I tackle this topic in a public forum; to my students, for having listened to me talk about this in the past and given me valuable feedback.

And second, I’d like to start with some caveats.  In particular, I am just an economist.  I’m not an evolutionary biologist, or a palaeontologist, or an anthropologist, or even an economic historian … just an economist.  But economics and economists are famous, and sometimes infamous, for not letting their lack of expertise stopping them from addressing topics outside their home turf, and that’s what I’ll be doing.

Finally, I need to start with an apology.  The advertised title of my talk is “The Economics of the Very Long Run: From Fire to Finance in Two Million Years”.  I chose the title over a year ago,  and while I’m still happy with the main title, the subtitle is a problem.  “Fire” is fine, because my starting point is 2 million BC and there is some recent work suggesting that fire figures here.  But the problem with “Finance” is that it implies an endpoint that is thousands of years too early.  So my apologies in advance for misleading you, and implying that I will have a lot to say about finance.  I won’t.  Fire, though, is a different story.

My endpoint is today – the era of modern economic growth that started with the Industrial Revolution two centuries ago, and continues today.  My starting point is, roughly speaking, the starting point of the human race, the emergence of the first members of our species in Africa about 2 million years ago.

The Big Picture
For 2 million years, until 1600-1800 or thereabouts, we can characterise the economic history of the human race very simply:

Population grows very slowly.
Total income – if you could value total goods and services produced by everyone, this is what it would be – also grows very slowly.
In fact, population and total income grow at essentially the same rate.
The result: average living standards, income per capita – is basically constant.
This is the “Malthusian Era” and it lasted for 2 million years.

The Industrial Revolution starts around 1600-1800 or thereabouts, depending on your perspective and what you think the causes are. I’ll say 1800 as a shorthand.

The Post-Malthusian Era – the era of modern economic growth – is very, very different from the two million years that preceded it.

The Industrial Revolution starts in England, spreads to the rest of Europe and portions of the New World, etc.  I’ll say “the West” as a shorthand.

In the West, there is rapid growth in income.  Population initially grows as well, but not as rapidly.  Then, population growth declines and population levels start to stabilise.  But income keeps growing.

The result in the West is rapid and sustained growth in living standards.

But in most of the rest of the World – I’ll say “the Rest” as a shorthand – income growth is slow and is matched by population growth.
The result in the Rest is population growth the continuation of the Malthusian Era living standards.

The result: the “Great Divergence” (Pomerantz).  The West gets fabulously rich, the Rest stays poor.  A few countries leave the Rest to join the West, but they are the exceptions.

The outline for the rest of my talk is as follows.  First, I will discuss the key features of the Malthusian Era in a little more detail, and sketch out the Malthusian Model.  Next, I will discuss two key developments during the Malthusian Era – the starting point 2 million years ago, when Homo erectus first appears and we cease to be upright apes, and the appearance of agriculture after the end of the last Ice Age.  I will then turn to the two centuries of the post-Malthusian Era and modern economic growth.  Finally, I will try to tie it all together, in the form of some hypothetical questions: why did this break in economic history – between the Malthusian Era and modern economic growth take place?  Was it inevitable?

Population and Living Standards: The Malthusian Era vs. the Modern Era
Brad DeLong has assembled some numbers on population, total income ( “GDP”) , and income per capita, i.e., living standards, going back to 1 million BC.

The picture for the Malthusian Era – through about 1600-1800  – is as I have described it already.  Population growth was positive but very, very slow: on the order of 0.05% per year, roughly the equivalent of doubling every two millennia.  Total income grew at the same rate as population.  And since total income and population grew at the same rate, income per capita was roughly constant.  DeLong’s data rely on work by Maddison, who uses 1990 dollars for his valuations.  In 1990 dollars, average living standards per capita in the Malthusian Era were about $400 per person per year.  Of course, there were fluctuations, but while these fluctuations may have been big compared to the average Malthusian standard of living – a couple of hundred dollars is a lot compared to $400 – but they were tiny compared standards of living we see today.

With the start of the Industrial Revolution, population growth and income growth both accelerate.  In the West, income growth outstrips population growth, and population growth accelerates but then slows down.  The result today is that population levels in the West are close to stable, but income continues to grow at 2% or so per annum.  In the year 2000, income per capita in the West was about $23,000 in 1990 dollars – about 57 times higher than the Malthusian Era level.  In the Rest, there is a mixed picture because we have a continuum of countries, some of which are becoming rich and appear to be joining the West, ranging all the way to some that are still desperately poor and where average living standards are not much different from the Malthusian Era level.

Of course, there are huge measurement problems here.  But for the most part, if we try to take these problems into account, they don’t change the picture.  And the big measurement problem works in our favour, as we shall see shortly.

Although I have presented these data in terms of population and income, and income per capita as calculated from the two, in reality the older projections backward are estimations of population and average living standards, and total income is the product of the two.  So if you have doubts about total GDP for 1 million BC, it’s not a problem.  The picture for 1 million BC of a very small world population of hunter-gatherers with living standards not much different from recent and historically-observed hunter-gathers is probably pretty accurate.

Another problem is that these living standards estimates are typically based on estimates of “material living standards”.  Of course, man does not live by bread alone, and art, music, etc. should figure in here somehow.  But there is little we can do about this.

The big measurement problem is a familiar one from index number theory: the “new goods” problem.  When we value output or consumption of goods for two different periods, which prices should we use?  The prices for the later period or the prices for the earlier period?  If the two periods are not far apart – say a few years – then it’s not a big problem.  But if the periods are a few decades apart, we start running into problems.  And if the periods are a few centuries apart, and one of the periods is today, then we get big problems.

The smaller version of the problem is that new goods are expensive when they first appear, but technological progress means they typically get cheaper over time.  I am old enough to remember the arrival of the first digital watches.  The first Pulsar watch sold for thousands of dollars in the 1970s.  Today, a cheap digital watches cost roughly a thousandth of this.  If we value today’s output and consumption of digital watches at today’s prices, it will be a thousandth of what someone in the 1970s would have valued it.  From the perspective of someone in the 1970s, we are richer than we, today, might think.

The bigger version of the problem is that digital watches didn’t exist before the 1970s.  In fact, a couple of centuries ago, most goods today didn’t exist at all.  What would someone from 1800 or 1900 have paid for things we treat as trivial and cheap: not just digital watches, but smartphones, antibiotics, an internet connection…?

The good news from my perspective is that this just accentuates the difference between the Malthusian Era and the Post-Malthusian Era of modern economic growth.  We are even more fabulously rich compared to our ancestors than the DeLong-Maddison data suggest.  DeLong has constructed a rough-and-ready theory-based approximation to take account of the “new goods” problem.  The effect is decrease the estimated level of average living standards in 1990 dollars by a factor of about 4, i.e., to roughly $100 per person.  So instead of being merely fabulously rich compared to our ancestors, we are whatever 4 times “fabulously rich” is.

The Malthusian Model
I think  I have made the case for trying to look at the Malthusian Era as a single era.  The prism through which I want to look at this period is, not surprisingly, the Malthusian Model.

From Thomas Malthus, “An Essay on the Principle of Population” (1798):

“[T]he increase of population is necessarily limited by the means of subsistence [and] population does invariably increase when the means of subsistence increase…”

This is a powerful idea.  Malthus is arguably the first modern growth theorist, the father of modern population ecology, and the grandfather of the theory of evolution.  It is ironic that he formulated his theory just as the world was beginning to leave the Malthusian Era.

The basic Malthusian Model has three components:

Births approximately equal deaths.  A population that is “in equilibrium” in this way will be approximately constant for a given environment.  The environment determines what this equilibrium population is.  To borrow a term from population ecology, the equilibrium population is the “carrying capacity” of the environment.

Say the population is below the carrying capacity of the environment.  Then food and resources are plentiful, mortality is low and life expectancy is high, and births exceed deaths.  The population will then increase until it hits the equilibrium level.  At that point, food and resources are no longer so plentiful, mortality is higher and life expectancy lower, and births equal deaths.

Say the population is above the carrying capacity of the environment.  Then food and resources are in scarce supply, mortality is high and life expectancy is low, and deaths exceed births.  The population will then decrease until it hits the equilibrium level.  At that point, food and resources are more plentiful, mortality is lower and life expectancy higher, and births equal deaths.

If the carrying capacity of the environment increases permanently – say there is some new technology invented, or new lands discovered – then population increases to match the higher carrying capacity.

How is this different from a population ecology model applied to some single species?

The Malthusian Model has 3 components: Equilibrium, Stability, Growth.  The first 2 components are the basic components of a simple population ecology model.  In a simple population ecology model, equilibrium is where births equals deaths.  If the system is perturbed or pushed away from equilibrium, the population of the species in question tends to return to this long-run equilibrium.  This could be a model of deer and forests, grey squirrels and nuts, or whatever.  In fact, it is common to cite Malthus as the first to formalise these basic principles of population ecology.

What is different – sort of – about the Malthusian Model is the third component, Growth.  Malthus explicitly addresses this, and the model wouldn't be a growth model if it didn't incorporate it.  In Malthus’ formulation of the Malthusian Model, the carrying capacity of the human environment increases at a slow rate through bringing new lands under cultivation and introducing new methods of agriculture, thus increasing the carrying capacity and hence the equilibrium population.  The stability features of the model derive from demographics – birth rates and death rates – and ensure that the actual population will reach the carrying capacity of the environment.

How is this different from models of evolution and natural selection?

The resemblance is more than coincidental!  Here is what Darwin says in his autobiography:

“In October 1838, that is, fifteen months after I had begun my systematic enquiry, I happened to read for amusement Malthus on Population, and being well prepared to appreciate the struggle for existence which everywhere goes on from long-continued observation of the habits of animals and plants, it at once struck me that under these circumstances favourable variations would tend to be preserved, and unfavourable ones to be destroyed. The result of this would be the formation of new species. Here, then, I had at last got a theory by which to work…”

And lest you think this is just a coincidence … Charles Darwin and Alfred Russel Wallace independently developed the theory of evolution.  Here is what Wallace writes in his autobiography:

“One day … I thought of [Malthus’] clear exposition of "the positive checks to increase" - disease, accidents, war, and famine - which keep down the population of savage races ... [it] occurred to me that these causes … are continually acting in the case of animals also …

It occurred to me to ask the question, Why do some die and some live? And the answer was clearly, that on the whole the best fitted live. … Then it suddenly flashed upon me that this self-acting process would necessarily improve the race, because in every generation the inferior would inevitably be killed off and the superior would remain - that is, the fittest would survive.”

So how is the Malthusian Model different from models of evolution and natural selection?

The Malthusian Model has 3 components: Equilibrium, Stability, Growth.  The first 2 components are shared with the Darwin/Wallace model of evolution and natural selection.  What is different is how long-run change takes place.

In evolution, it is natural variation – mutation etc. – combined with natural selection that provides the dynamics.  A new mutation arises that provides an individual member of the species with greater “fitness”.  Because of this, the individual has more offspring than other members of the species, and the new characteristic spreads through the population.  Eventually the new characteristic is seen in all members.  If the characteristic enables individuals to obtain more resources from a given environment, or to better survive the stresses it faces (predation, disease, whatever), then the equilibrium population will be higher.  This needn’t happen, by the way; the new equilibrium population could be lower or higher depending on, say, whether a mutation causes reproductive strategy to raise or lower birth rates.

In the Malthusian Model, human activity itself moves the stable equilibrium population.  This human activity can be, say, bringing new land under cultivation – Malthus' example.  But it can also be new ideas, tools, methods, inventions, organisations … all of these innovations change the environment and its capacity for sustaining human populations.  And the spread and replication of these ideas – “memes” (Dawkins) – does not require genes.  Individuals and groups learn from and copy each other, and new ideas spread through the population as a result.

This last point – the role of innovation and new ideas in generating economic growth – is one to which I will return at several points.

In the lecture, following Clark (2007), I set out a simple graphical version of the basic Malthusian Model.  The presentation has two graphs.  The equilibrium and stability features of the model are shown in a graph of birth and death rates vs. per capita living standards y.  At high living standards, births exceed deaths.  At low living standards, deaths exceed births.  Where the birth and death rate schedules cross tells us living standard of living y* at which births and deaths are equal and the population is in equilibrium.  The second graph shows the relationship between average living standards and the total population for a given technology and resources, where “technology” means knowledge, tools, etc., and “resources” means land and everything else in the environment.  This “technology schedule” or “productivity schedule” captures the feature that for a given environment, a larger population will have lower average living standards.  The economics jargon is “declining marginal product of labour”, but the intuition isn’t hard to see.  If a hunter-gather or farming community experiences a doubling of its population, it will be forced to bring into use marginal, low quality land, or to eat foods it wouldn’t normally use, etc.  An innovation such as fire changes the technology schedule so that the same resources can support a larger population.

Which brings me to…

Homo erectus, fire and cooking: When did we become human?
The genus Homo appears about 2 million years ago with Homo erectus.  Prior to that we were

Australopithicines and habilines (early Homo): morphologically ape-like but with larger brains.  We were tool-users and these moderately larger brains may have been associated with increased meat-eating.  But we still more morphologically more ape-like than human.  The break between habilines and Homo erectus is a big one, involving major morphological changes: increases in body size; loss of features associated with climbing; smaller teeth, jaws, guts (digestive systems).  And we become much more geographically widespread – Homo erectus is the first hominid to leave Africa, and was found as far as China and south-east Asia.

The hypothesis advanced by Richard Wrangham and his coauthors is that these major changes are the result of a new technology – fire.  And the implications go well beyond these morphological changes – this may be when we also become behaviourally human.  I should note that the Wrangham hypothesis is new and controversial, and not (yet?) generally accepted.

Chimpanzees and bonobos, like early humans and pre-humans, are hunter-gatherers.  They forage for plant foot and also hunt and eat meat, though somewhat less of the latter than humans.  The key difference between chimpanzees and ourselves is that they eat their food raw.  Raw food is very hard to digest; it takes a long time to chew, and a long time to pass through the gut.  The evolutionary consequences are visible in chimpanzees and australopithicines: large jaws, large teeth, long guts.

Cooking is an extremely effective mechanism for pre-processing food.  It makes food much easier to digest, nutritional content more easily absorbed.  Modern humans have much smaller jaws and teeth, and much shorter guts (intestines), than chimpanzees.  Modern humans find it extremely difficult or impossible to survive on a diet purely of uncooked food.  All known human societies cook their food.  Wrangham suggests that it is the invention of fire of cooking 2 MYA that led to the emergence of Homo erectus.

The Wrangham hypothesis is speculative, and the big gap in the argument is the absence of direct evidence for fire use 2 MYA.  The oldest evidence so far is about 1 MYA, published less a year ago (i.e., after Wragham’s writings – a point in his favour).  His main evidence is indirect – it is difficult to impossible for morphologically modern humans to survive purely on raw food.

Wrangham argues that fire and cooking brought with it a large number of morphological and behavioural changes, changes that “made us human”:

Shorter guts, smaller teeth and jaws, larger body size indicative of a better/more abundant diet.

Loss of climbing abilities.  Fire is a weapon, can protect against predators, especially at night.  Homo erectus doesn’t need to climb trees for self-protection.

Larger brains.  Brain size increased from about 600 cm3 in habilines to about 900 cm3 in Homo erectus.  (Australopithicines – about 450 cm3; modern humans – about 1400 cm3.)  The human brain is an extremely energy-intensive organ.  In modern humans, it accounts for 2.5% of body weight and 20% of energy use.  The evolutionary importance of this was pointed out by Aiello and Wheeler in their “Expensive Tissue Hypothesis”.  Wrangham’s answer is that cooking enabled the evolution of big, expensive brains.

Hairlessness, an unusual feature for an ape.  Wheeler: hairlessness has a major advantage in hunting – humans do not overheat nearly as quickly as fur-covered animals.  Human hunting strategy frequently involves chasing down an animal until it is exhausted, i.e., overheated.  The cost of hairlessness is being cold at night.  Not a problem if fire technology is available.

Reduction in sexual dimorphism, and the emergence of pair-bonding and modern human sexual patterns.

This last point is the most speculative, but the most appealing to an economist, because it involves the applications of game theory to an investment problem.  The argument is that cooking takes time, and while it is being cooked it can be stolen.  This is how chimpanzees behave – chimp society is hierarchical, and those high in the hierarchy steal from those lower down.  Human hunter-gather society is more egalitarian.  Moreover male-female pair bonding is the usual pattern, and in the sexual division labour in almost all observed human societies, industrial and pre-industrial, women do more cooking.  Wrangham argues male-female cooperation emerged via a repeated game in an evolutionary context.  Females gather food and cook, and individual females form pairwise cooperative arrangements with individual males.  While female cooks, male helps guard food.  When food is ready, both consume.  These alliances are lengthy – they are not just for today’s meal.  The reasoning is that pair-bonding creates children who are a joint investment into which it is profitable for both parents to invest.  We can contrast this with promiscuous chimpanzees – the fathers don’t know who their children are – and with animals where, when a new male arrives and takes over, he kills all the existing young, because he knows who his children aren’t.

This is, admittedly, a somewhat speculative hypothesis, and Wrangham acknowledges this.  Probably the main competitor also builds on the Aiello-Wheeler “The Expensive Tissue Hypothesis” but argues that the main breakthrough was not cooking, but meat-eating.  Meat is very high quality food.  In this perspective, meat-eating (and implied tool/weapon use) was the key early step, and cooking came later.

My perspective on the cooking and fire argument is that we can see it through the prisms of both the Malthusian Model – economics – and the Darwinian Model – natural selection.  From the Malthusian perspective, it was a huge technological advance with major social and behavioural consequences.  From the Darwinian perspective, it also had major evolutionary impact on humans, which in turn meant morphological, social, behavioural consequences.  The result, from the Malthusian perspective, is significant population growth and spread of Homo out of Africa.

We have a problem here that is similar to the “new goods problem” but even worse.  How can we compare living standards of different species?  Habilines and early humans would not have put much value on digital watches.  Still, it is probably reasonable to conclude that fire and subsequent technological advances over the next million years or so translated primarily into population growth and not improvements in living standards.  In other words, the basic Malthusian Model was in operation.  I will return later to where these technological advances came from, and how often they appeared.

Agriculture is the most important technological development between the emergence of Homo 2 million YA and the Industrial Revolution.  It first emerged in Middle East after the end of the last Ice Age.  It subsequently spread to other ancient societies in Old World, but was also separately invented multiple times in the Far East and the Americas.

The history of early agriculture one of the appearance and spread of innovations: first the cultivation of cereals, other plants; selective breeding (“genetic modification”); domestication of animals: dogs, pigs, cattle, goats, llamas, etc.

All this makes possible large increases in numbers and population density, because the same land can support a much larger agricultural population than hunter-gatherer population.  Over the millennia, the agricultural lifestyle spreads and either pushes out or absorbs neighbouring hunter-gatherers.  The genetic evidence from Europe is that 20-30% of European DNA derives from immigrant farmers from western Asia.

Agriculture made possible major transformations in human society:

Sedentary lifestyles.  People live in one place continuously.
Physical capital accumulation possible; large tools but especially buildings.
Urbanisation, towns/cities.  A large agricultural hinterland can support a central town.
Hierarchical societies.  Aristocracies, priesthoods, wealth and income inequality all emerge with agriculture.
City states and nation states.

And here I have to say something about finance because of my mis-chosen title.  Finance, writing, accounting, debt, etc. emerge with hierarchical societies.  These are major innovations, but not as fundamental as agriculture, nor do they change the Malthusian Equilibrium.

… Which indeed is essentially unchanged.  Agriculture enables a substantial increase in population: from about 4 million people worldwide in 10,000 BC shortly after it first appeared, to about 50 million worldwide in 1000 BC, and about 170 million worldwide in 1AD.  This represents a huge increase in population growth rates compared to the miniscule pre-agricultural era growth rate of 0.0004% per annum.  Average world population growth was about 0.04% per annum between 10,000 BC and 1 AD, about 100 times faster, driven by a combination of the displacement of the hunter-gather-lifestyle by that of agriculture, and improvements in agricultural techniques and other technologies.  But although the world population grew hugely, there was no major change in material living standards.  Indeed, if anything, they probably fell, as we’ll see in a second.

A simple application of the Malthusian Model would say that this is best interpreted as the natural consequence of an outward movement of the technology schedule.  For a given population, agriculture enabled a higher standard of living, leading to a lower death rate and hence population expansion.  But recent work by Bowles (2011) suggests the picture is more complicated than that.

First of all, it is not at all obvious that from the perspective of individual hunter-gatherers, farming was a better way to make a living.  Bowles, Diamond and others have pointed out that agriculture was a health and nutrition disaster, and hunter-gatherer diets are much healthier than those in agricultural communities.  Agricultural societies rely on unbalanced high-carbohydrate diets, heavy reliance on a few foods.  This is poor nutritionally, and exposes farmers to greater risk (just one crop failure and you are in big trouble!).  Hunter-gatherers have wider variety (and this is what we evolved to survive on).  Agriculture is also associated with disease – many of the major diseases that we suffer from as humans originated as diseases in domesticated animals and jumped the species barrier.  All this shows up shows up in archaeological record: lifespans, disease, heights, etc.  Average stature collapsed with the introduction of agriculture; members of agricultural societies were – and often still are! – very short.  Modern Europeans are only now returning to the heights of Ice Age Europeans.

Evidence assembled by Bowles shows, moreover, that labour productivity by the earliest farmers was probably lower than that foraging for wild species, i.e., in hunting and gathering.   If so, then why was it adopted?  Bowles points out that his estimates are of average productivity, and the optimal answer to the decision problem facing these early proto-farmers would have been based on marginal productivity.  In the jargon of microeconomics, early adopters of agriculture would allocate some time to foraging and some time to farming until the marginal product of labour is equalised.  As Bowles notes, however, this just moves the question ahead one step: why did some early populations become mainly farmers?  If a group became mainly farmers and abandoned foraging, then the average productivity would have been low compared to groups that remained hunter-gatherers.  They would be better off abandoning this specialisation in farming and going back to partial or full foraging.

Bowles’ speculative answer is basically demographics.  It is likely that for a few groups, the conditions they faced made it sensible to adopt farming as their main activity.  And once they adopt farming, living in a fixed location means the cost of child rearing falls.  Hunter-gatherer groups face serious constraints on birth rates – babies are a major burden for groups that regularly have to travel large distances.  Thus the causality runs from early agriculture/fixed location => higher birth rates => population growth => more land needed for additional agricultural communities => spread of agriculture.  In this perspective, what led to the spread of agricultural was the combination of increasing birth rates and a movement of the technology schedule.

The Era of Modern Economic Growth
The era of modern economic growth start with the Industrial Revolution, and as a shorthand I will date this as 1800.  In the West, where Industrial Revolution started, and modern economic growth began, we have a fairly good understanding now of how it works.

The key role in modern economic growth is played by invention and innovation, technological and organisational, a perspective often identified with the great 20th century economist Joseph Schumpeter.  “Growth accounting” exercises show that nearly all growth in income per capita and living standards can be attributed eventually to technological progress.  This growth is generated by a substantial R&D sector – a significant portion of the population is engaged professionally in the creation of new knowledge.

A small-ish number of countries from “the Rest” have joined the West, e.g., Japan in the 19th and early 20th century, South Korea in the second half of the 20th century.  A somewhat larger but still small number of countries are – probably – in the process of joining the West.  The process of economic growth in these countries is also fairly well understood, and follows the pattern of rapid “catch-up growth” noted by Veblen and Gerschenkron.  What Gerschenkron called “the advantage of backwardness” is that countries can temporarily grow very quickly by adopting the established world-best technology – there is no need to start with Watt’s steam engine and then gradually progress to modern turbines.  Eventually these countries, once they finish catching up,  slow down and grow at the same rate as the rest of those at the technological frontier.

And many countries still have not, or haven’t yet, started the catching-up process.  These countries are still desperately poor.  They see slow growth of total income that is close to being matched by population growth.  These growth rates are fast compared to the glacially-slow growth rates of the Malthusian Era, but very slow compared to the growth of total income and living standards in the West.

Why so many poor countries are still poor, and have not made the breakthrough to modern economic growth, is, however, not well understood, or not well enough at any rate (as the history of aid policies to the developing world over the last 50 years attests).  There is a vast literature here and something I do not want to even try to address here.

What I do want to discuss in the time remaining is the role of demographics in the emergence of modern economic growth, how it relates to the nature of innovation and technological progress – the generation of memes – and to engage in some speculation about the Industrial Revolution and historical inevitability.

Demographics played a key role in break between the Malthusian Era and the modern era.  In the West, after the start of the Industrial Revolution, income started growing rapidly, and population grew too.  This is what the Malthusian Model would predict.

But then something happened that meant a break with the Malthusian Era: population growth in the West slowed down, and this despite technological advances and increases in living standards that would have reduced mortality.  What happened was a fall in birth rates – in fertility – that was extraordinary in scale.  In 1820, the fertility rate in England was in excess of 50 births per 1,000 population.  By 1900, it was below 30.  Today it is close to half that again.  The key break with the Malthusian past is essentially demographic: self-sustaining economic growth now translates into growing living standards and stable populations.

Why did birth rates fall?  A huge question, and one that I again don’t have time to go into.  I will mention only a few points.  The first is modern methods of contraception, which start to appear in the West in the second half of the twentieth century.  The second is the opportunity cost of having children, especially for women.  Industrialisation means an increasing role of women in the workforce, and there is some evidence that these improved opportunities for women led to reductions in fertility.  The third is the nature of investment in children.  It has been suggested that because of developments in education and the need for skills and training, parents voluntarily reduced fertility in order to be able to invest more in a smaller number of children – a quantity-quality trade-off.

Population size and population growth is also important for understanding the generation of knowledge.  New innovations and inventions today generate rapid economic growth not only because of the increase in the numbers of people engaged in innovative and inventive activity compared to the Malthusian Era, but also because of the nature of knowledge.

Knowledge is special: when something new is invented, the idea can be “reproduced” costlessly.  One group invents the bow and arrow, and it spreads through the entire population.  The modern economics jargon for this is “non-rivalrousness” – ideas can be consumed by one person without reducing the availability of the idea for others.  This point was known in Malthus’ time.  Thomas Jefferson made it very well in 1813 (HT: Timothy Taylor), and his jargon is much nicer than ours today:

“If nature has made any one thing less susceptible than all others of exclusive property, it is the action of the thinking power called an idea, which an individual may exclusively possess as long as he keeps it to himself; but the moment it is divulged, it forces itself into the possession of every one, and the receiver cannot dispossess himself of it. Its peculiar character, too, is that no one possesses the less, because every other possesses the whole of it. He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me. That ideas should freely spread from one to another over the globe, for the moral and mutual instruction of man, and improvement of his condition, seems to have been peculiarly and benevolently designed by nature, when she made them, like fire, expansible over all space, without lessening their density in any point, and like the air in which we breathe, move, and have our physical being, incapable of confinement or exclusive appropriation.”

An important implication of this is that the speed at which technology advances increases as the population grows.  The more people there are in the world, the more thinking and experimenting is done, and the more ideas are generated.  Once good ideas are created, they spread – “costlessly”, in the sense I’ve just mentioned – throughout the population.

This is one way how human history differs from other species: genetic change does not have the same tendency to accelerate with population size.  Increases in knowledge (“memes”) do not face the same constraints as genetic changes.

This positive relationship between the size of the world population and the speed at which knowledge advances is visible in the evidence we saw earlier from the Malthusian Era.  Population growth in the Malthusian Era gradually accelerated over time.  Another bit of interesting evidence relates to the possibility of technological regress – small isolated populations of hunter-gathers have sometimes moved backwards technologically.  During the Ice Age, Tasmania and Flinders Island, a small island between Tasmania and mainland Australia, were linked by a land bridge.  After the rise in sea levels at the end of the Ice Age, Tasmania and Flinders Island were separated.  The small-ish population of Tasmanians experienced technological regress compared to their mainland Australian relatives, and some technologies were lost.  The Flinders Islanders … went extinct.

The break with the Malthusian Era in terms of the creation of new knowledge is not, of course, just a mechanical connection with the large populations of the 1600-1800s compared to earlier era.  There was a fairly sharp break that translated into a jump in growth rates of income and then income per capita, and all that accompanied it.  Why?  Why then, whey there?

This too is hugely debated, and a wide range of arguments have been proposed.  Again, there is no time to go through them in any detail; I barely have time to mention some of the contenders:

Institutions: property rights, limited government, etc.
The co-location of coal and the incentives in the form of high wages to use it to substitute energy and machinery for labour.
The Enlightenment and the emergence of a substantial innovative sector – persons engaged in science, engineering, etc.
Even genetics and cultural transmission of values (Clark’s argument about fertility patterns in England – rich commoners had more offspring).

But frankly, we really don’t know.  The more interesting question for me, and one where I hope I do have some to talk about, is: was the Industrial Revolution inevitable?  Of course, this question is unaswerable, but it is hard to resist the temptation to speculate.  I will give in to temptation completely, and ask this question about the Industrial Revolution in science-fiction-alternate-history (“Strossian”) terms.

What if, say, England were coal-less?  Or perhaps the era of Classical Antiquity hadn’t ended and the Western Roman Empire had survived and assimilated the latest wave of immigrants?  I can envision two different scenarios:

Scenario 1: Population growth and increasing returns mean that somewhere in the world (China?) a tipping point is reached, modern technology-led growth starts up, population initially grows rapidly and then slows down, modern economic growth starts.  A parallel universe that is actually parallel.

Scenario 2: No tipping point.  Knowledge continues to accumulate, but slowly.  Population growth never slows down.  Living standards stay at the Malthusian level.  To borrow Douglas Adams’ terminology, this is a strange un-parallel parallel universe: agrarian in the distribution of income – most people extremely poor, an elite that is very rich – but still gradually accumulating knowledge and technology, and perhaps not very different from the “the Rest” of the world today.

But I think we can actually say something concrete if we change the hypothetical: what if, also, the world were coal-less?  Or, more precisely, solid-fossil-fuel-less?  Solid fossil fuels have a huge energy density – energy per kilogram – and are extremely abundant compared to the preceding alternatives (wood burning, water wheels, whatever).  How important was energy in the Industrial Revolution?  Tony Wrigley and others have argued that the answer is “Very Important”.  I agree.

Scenario 1 seems much more unlikely in this hypothetical world – energy seems to be a crucial ingredient in generating the tipping point.

Scenario 2 seems much more likely.  We would eventually have a world of “highly-developed” (mathematics, optics, music and art, …) agrarian economies with extreme inequality.


© Mark Schaffer 2013



References and Further Reading

Leslie C. Aiello and Peter Wheeler (1995), “The Expensive Tissue Hypothesis: The Brain and the Digestive System in Human and Primate Evolution”, Current Anthropology, Vol. 36, No. 2, pp. 199-221.

Samuel Bowles (2011), “Cultivation of Cereals by the First Farmers was Not More Productive than Foraging”, PNAS (Proceedings of the National Academy of Sciences), Vol. 108, No. 12, pp. 4760-5.

Gregory Clark (2007), A Farewell to Alms: A Brief Economic History of the World, Princeton University Press.

J. Bradford Delong (1998), “Estimating World GDP, One Million B.C. – Present”.  Available online at

Jared Diamond (1987), “The Invention of Agriculture: The Worst Mistake in the History of the Human Race”, Discover magazine.

Timothy W. Guinnane (2011), “The Historical Fertility Transition”, Journal of Economic Literature.

Kenneth Pomerantz (200), The Great Divergence: China, Europe, and the Making of the Modern World Economy, Princeton University Press.

Charles Stross (2004-2010), Merchant Princes, TOR Books (multiple volumes).

Richard Wrangham (2009), Catching Fire: How Cooking Made Us Human, Basic/Profile Books.

E.A. Wrigley (2010), Energy and the English Industrial Revolution, Cambridge University Press.