K.G.Persson
Note 2. Population, Economic growth and Resource constraints.
Historical trends in population growth
Economics is sometimes called the dismal science because many of its pioneers expressed
pessimism about the possibility of sustained economic growth in a world of limited resources. We
meet this view today in the worries about raw material shortages, such as oil, eventually putting an
end to economic growth. Late 19th century economists worried about coal shortage, but the concern
today is rather that coal generates too much CO2 which might in the long run harm growth. The
first economist to develop a coherent theory of limited resources as a binding constraint for
sustained long run economic growth was Thomas Malthus whose Essay on Population was first
published in 1798. The date of publication is not without interest since it coincides with the first
decades of the Industrial Revolution, which combined unprecedented population growth with
increasing (or at least constant) income per head. Malthus argued that given limited land the supply
of food would eventually constrain income and population growth. We will soon return to a detailed
exposition of the Malthusian view but we will first review the evidence on long run population
growth in Europe and elsewhere. Reasonably precise population estimates are available only from
the 16th century and before that populations are estimated from projections based on scarce data and
conjectures of the carrying capacity of a given surface at the prevailing technology.
Cultures based on hunter-gatherer technology, which preceded the breakthrough of agriculture and
sedentary civilization in the Middle East some 12.000 years ago, were very demanding in terms of
land requirement which limited world population to an estimated 6- 8 million. A hunter-gatherer
culture satisfies its food requirement from nature without actually controlling it: as a consequence
there will evolve an equilibrium size of the stock of animals and the stock of men. If man
overexploits the animal stock the reproduction of both will be disturbed. As the animal stock
declines the size of human population will also fall. Technological progress in a hunter gatherer
culture, say better arrowheads, will therefore not spill over permanently into higher food
consumption or a larger population but only in more leisure. The high proportion of leisure in
surviving hunter and gatherer cultures has sometimes been romanticized by anthropologists, but is
actually imposed by the peculiar character of hunter gatherer technology, that of living from nature
rather than by deliberately controlling it.
The Neolithic revolution*, that is the transition from hunter gatherer cultures to agriculture,
introduces a growing element of control of nature and gradually reduces the land requirement
1
needed to feed a man. As a consequence agricultural civilizations are accompanied by a permanent
growth of world population. From an estimated 6-8 million at the dawn of the Neolithic revolution
world population increased to some 250 million at the beginning of CE (Common Era), one billion
when Malthus published Essay on Population and six billion in 2000 CE.
Graph 1a and 1b. Population Growth 400 BCE to 2000 CE.
500
450
Asia
Europe
400
America
Population in millions
350
300
250
200
150
100
50
0
-400
0
400
800
1200
1600
4000
America
3500
Europe
Population in millions
3000
Asia
2500
2000
1500
1000
500
20
00
19
80
19
60
19
40
19
20
19
00
18
80
18
60
18
40
18
20
18
00
17
80
17
60
17
40
17
20
17
00
0
Sources: Graph 1a: J-N. Biraben, Essai sur l’évolution du nombre des hommes , Population , vol 34, No. 1, 1979, pp.
13-25. Graph 1b: Netherlands Environmental Assessment Agency,
http://www.mnp.nl/hyde/Images/pop_summary_tcm63-22929.xls
2
Graphs 1 a and b portrait population growth over the last 2400 years. The first graph smoothes the
actual variations because the curves are based on estimates given in intervals of 100 to 200 years. In
between the estimation points political crises, epidemics, harvest failures and wars had severe
effects on population. However, graph 1a catches the major characteristics of pre-industrial
European population as slow trend growth interrupted by two major setbacks. The first is associated
with the decline and crisis of the Roman Empire which was triggered off by internal political
conflicts, invasions, massive migration flows and recurrent epidemics. Between 200 CE and 600 CE
Europe’s population might have been reduced by half until it started to grow again. The next
setback is again linked to an exogenous shock, the Black Death in 1347, a highly contagious pest
called yersina pestis carried by fleas using rats as their host. The plague was not a single event but
ravaged Europe and Asia for about 100 years reducing the population in Europe by a third. Europe
did not attain its pre-plague population until the mid 16th century.
China has more reliable population data for the pre-industrial period but the general pattern of
positive trend growth interrupted by exogenous events such as invasions and war, epidemics and
political disintegration, is supported by the data. Apart from the large exogenous shocks to
population there are also short run shocks, driven ultimately by variations in harvest outcomes,
which in turn were largely influenced by unforeseen climatic shocks or plant diseases. A poor
harvest was often associated with epidemics because hard times triggered off migration which
carried diseases.
The second graph covers a period when Europe is undergoing its ‘demographic transition’ from
slow growth in the early 18th century to a faster growth by the end of that century lasting a century
or more until the slowdown of population growth in the present period. If we concentrate on the
two hundred years since the publication of Malthus’ influential book the fears that he expressed
seem to be utterly misplaced. Population has increased by a factor of six, world food production by
a factor of 10 and still not all land fit for agriculture is currently used. According to Italian
economic historian Giovanni Federico between 80 to 90 percent of all cultivable land is now used.
However, that Malthus was wrong about his own period and the future need not imply that he was
wrong about the past. In fact his analysis is routinely invoked by economic historians and
economists as a meaningful model explaining pre-industrial economic history, so we need to take a
closer look at the Malthusian view.
3
The Malthusian theory of population growth and stagnation.
One reason for the continuous appeal of Malthus’ view is its seemingly irresistible logic and
simplicity. However, as we will see, logic does not necessarily imply historical relevance.
There are three basic tenets: population growth is positively correlated with income per head, land
is in limited supply and food supply can not keep up with continuing increase in population because
of falling marginal productivity of labour. The irresistible logic referred to above is that there is no
escaping from the fact that there are limits to the amount of cultivable land. As population is
increasing the land available for agriculture per head will fall, which at some point should lead to
falling marginal product of labour. But take care! The argument assumes implicitly that there is no
technological progress in agriculture, that is, there is no increase in the efficiency of the use of land.
As we will see, technological progress in agriculture is ‘land augmenting’: it is increasing the land
available for cultivation per year.
Graph 2: Malthus graphically speaking
Vital Rates
Constant, above
subsistence, income and
positive population
growth.
CBR
CDR
Malthusian Equilibrium
CBR-CDR=0
Income= Subsistence
Constant population
Time
Income per head
Land/Labour
4
Terminology Box
So called vital rates are Crude birth rates (CBR) and Crude death rates (CDR) and they are normally
measured per thousand and per year. CBR is referred to as fertility and CDR as mortality in the text.
Population growth per thousand is equal to CBR-CDR.
Positive check. If population is above its equilibrium level (or its equilibrium growth rate) a
temporary increase in mortality will permanently reduce population to its equilibrium level (or
equilibrium growth rate).
Preventive checks: cultural or moral restraints on fertility
Total fertility is the expected live birth per woman.
CBR is at its maximum when income is high because households need not worry about economic
constraints. Preventive checks, that is deliberate reduction of fertility, will set in as income falls,
however. In modern parlance we would say that Malthus assumed fertility to be very income
elastic. Demographic research suggests that falling income reduces the proportion of a generation
ever getting married and/or delays age of marriage thereby reducing fertility. CDR increases with
falling income because populations become more vulnerable to poverty related diseases and deaths
since the nutritional standard deteriorates with falling income.
Imagine now a sparsely populated economy where farmers have plenty of land, which implies that
labour productivity and income is high. Population growth, the difference between CBR and CDR,
is high and the increasing population will bring down the land/labour ratio. Income per head as well
as the rate of population growth decline until the Malthusian equilibrium is reached, which is
characterized by subsistence income and a constant population, that is CBR-CDR=0. In the absence
of technological progress the Malthusian equilibrium is an inevitable outcome.
Exploratory Box
Is the Malthusian theory testable?
Long lived theories paradoxically evade proper testing. Is that also the case with the Malthusian
theory? British economist Mark Blaug famously dismissed Malthus as ‘a perfect example of a
tautology masquerading as a theory’. This harsh judgment is based on the view that Malthus did
not specify a precise outcome that could be refuted or corroborated. Italian demographer Massimo
Livi-Bacci, however, proposes a long term Malthusian population trajectory along the lines
explored by Graph 2. Population will grow fast initially, only to see the rate of growth falling when
land scarcity sets in until a zero population growth rate is reached, which is the Malthusian
5
equilibrium described in Graph 2. If we take Livi-Bacci at his word we can safely dismiss the
Malthusian theory as a prediction of long term population growth. Graphs 1a and b indicate a
positive trend in population interrupted by large exogenous shocks. However, long run population
growth can be included into a Malthusian model if technological progress is admitted for. Blaug
was right in that there is ambiguity in Malthusian predictions depending on the choice of
assumptions.
Malthusian thinking proposes that long run fertility and mortality rates are determined by real
income but there is little evidence that this is the case. Appendix Figure 1 indicates for example, a
gentle rise in English real farm wages in the 17th century while the rate of population growth falls.
From around 1750 to 1800 real wages decline while population growth increases.
In the short run there seems to be stronger response of fertility and mortality to harvest shocks
which cause dramatic price increases, migration and diseases. Harvest shocks tended to trigger off
excess mortality in Europe at least until the end of the 18th century or early 19th century and there
are isolated cases of severe excess mortality due to harvest failures in Finland and Ireland as late as
the middle of the 19th century. It is not obvious that excess mortality in a famine can be interpreted
as a Malthusian positive check in the sense that it reduces population or population growth
permanently. In fact most famine mortality periods are followed by excess fertility and below
average mortality until the trend population growth has been restored.
Dissatisfied with the absence, in the long run, of the expected Malthusian demographic response to
economic variables such as real wages, Irish economist Morgan Kelly has looked at climate
variations and discovered slightly better explanatory power of climate (than of real wage) on
demographic responses, that is crude death and birth rates. The logic is that a deterioration of the
average temperature will affect agricultural production and hence living standards, which will have
an impact on fertility and mortality. However, these responses seem to appear in very different
epochs and at varying land/labour ratios, so it is not clear that it is a test of the Malthusian model.
The most articulate Malthusian among modern economic historians, Gregory Clark of UC Davis,
actually undertakes a test of the proposition that there is a negative relationship between real wage
and population density over time. An increase in population means that the land to labour ratio
falls and the Malthusian postulate, as demonstrated in Graph 2 is that income or real wage falls.
There is evidence of a trade off between population and real wages in pre-industrial England at
least until the end of the 16th century, Clark argues
6
Graph 3 Cross-section and time series relationships between real income and population
density in the pre-industrial period
Real income
Cross-section relationship in pre-industrial
Europe
1450
1200
1550
1600
Population density
The convex curve in Graph 3 illustrates Clark’s finding but we have also entered an observed
cross-section* phenomenon for pre-industrial European regions which suggests a positive
relationship between population density of regions in Europe and the regional real wage. The cross
section suggests that at a single point in time, say year 1300 or year 1700, densely populated areas
had higher labour productivity and income per head than less densely populated areas. Clark’s data
for England suggest that labour productivity falls or is stagnant during most of the pre-industrial
period. However there is independent evidence, to be reviewed in Note 3, that labour productivity
increases in England from the 17th century onwards. The question is of course why it is not
reflected in farm wage increases. Since both land and labour markets are far from perfect you
cannot make direct inferences on marginal product from wage data. Wages reflect the bargaining
strength of labour which is inversely linked to population density. When labour is scarce in the mid
15th century after 100 years of plagues, the bargaining position of farm labour is strong which
pushes up wages, while the ensuing growth of the labour force from then on improved the
bargaining power of landowners. Real wages are a good guide to labour productivity only if wages
are assumed to be a constant share of national income, which they are not.
7
The nature of technological progress in agriculture.
The historical and empirical support for the Malthusian model is not strong and we will now
explore the characteristics of the model which makes it problematic in a historical context, before
and after the Industrial Revolution.
There are three major problems in the Malthusian model: (i) it ignores or underestimates
technological progress, (ii) it characterizes the economy as closed, i.e. ignore the impact of regional
and foreign trade, (iii) it lacks a theory of fertility strategy for households based on optimizing
behaviour.
The secrets of agricultural progress.
The focus on land as a limiting resource ignores the fact that land, say a hectare of cultivable land,
is an extremely poor predictor of the yield of that land in the long run. In the history of agriculture
the crop ratio, which is the number of crops per year and unit of land has increased from a ratio of
0.05 in primitive slash and burn cultivation* to close to one crop per year in Europe (but higher in
horticulture) and a maximum of 2 in some regions outside Europe when it comes to cereals such as
rice. The crop ratio has in other words increased by a factor of forty in the very long run.
Furthermore the yield per unit of land and crop has also increased meaning that the combined yield
times crop per year and unit of land ratios has increased even more, say, by a factor of 150 in the
long history of agriculture. This is what technological progress is doing to agriculture.
Technological progress is based on the accumulation of useful knowledge which in pre-industrial
economies is growing slowly, often more by chance than by deliberate experimentation or trial and
error. However, needless to point out, the knowledge gained in learning by doing has no scientific
base. Pre-industrial producers could see that manure increased the yields but did not know that the
essential mechanism here was nitrogen.
In agriculture experience was linked to the management of soil fertility, mainly the maintenance
and release of nitrogen. The stock of nitrogen is only slowly released by mineralization, which
makes nitrogen in the form of ammonium and nitrates water soluble. You need to add directly
accessible nitrogen as well by means of manure. The nitrogen in grain fed to animals or men will be
recycled as manure. Soil fertility is also dependent on the choice and rotation of crops and the
extent of fallow, which is a period where the land is not used for tillage but used for cattle breeding
or left to rest. Some crops, for example clover, are better than others in fixing nitrogen in the air and
if included in a crop rotation made possible better use of land and an increase in yields. Other crops,
8
such as beans and peas, had a long term effect on yields of grain although the short run effect was
negative, because they increased the stock of nitrogen in the soil. The control of soil humidity and
ploughing technology are both essential for the release of nitrogen. Nitrogen was left in roots and
other residues of plants and needed to be decomposed to be of any use. Too much rain might wash
away nitrogen in the form of water soluble nitrates. Yields will also be dependent on the control of
competing plants, weeds, and the administration of animal and human manure, and more recently
chemical fertilizers. Increases in nitrogen will increase yields over a fairly large input range and the
response in terms of yields is very high, in the order of 15. So a one kg increase of nitrogen on a
unit of land increased yields by 15 kg. The supply of manure requires a balance between tillage and
animal husbandry which can be difficult to maintain if land is scarce. However, in densely
populated areas human manure, night-soil, from neighbouring cities can be a substitute for animal
manure. The stock of nitrogen in the soil is slowly released and will be depleted in the long run if
land was not properly fertilized or if the land was not left in fallow.
Land differs in soil characteristics as much as plants have different needs when it comes to location,
climate and soil characteristics. The complementarities of soil, climate and plants cannot be
exploited without trade and trade will contribute to better land use, which will affect yields
positively. By and large densely populated regions in pre-industrial Europe were characterized by
high agricultural productivity as is suggested in Graph 3 and it will be further explored in Note 3.
At first sight this seems to be a paradoxical finding. However, the proximity to urban centres gave
agriculture in densely populated regions a number of productivity enhancing possibilities. The
shortfall in nitrogen from land intensive cattle breeding could be substituted by human manure,
night-soil from the cities. New knowledge is also diffused more swiftly in densely populated areas.
The varied demand from cities permitted peasants to diversify into poultry, horticulture and
industrial plants, for example flax and plants used for dyeing. Furthermore a diversified crop
‘portfolio’ was an insurance against natural accidents since harvest failures often were specific to a
single crop or plant at a given point in time. Again trade helps in putting scarce land into best use
and the densely populated regions were usually dependent on cereal imports form other regions and
nations. Land and capital markets were much more developed in the proximity of urban centres
which contributed to getting land into its most profitable use compared to remote sparsely areas
with little access to consumer markets and little scope for diversification of crops.
9
If we return, for a moment, to Graph 2 we can spell out the implication of technological progress
and more efficient land use facilitated by trade. Malthusians have tended to focus on limited land
supply neglecting the fact that output per unit of land is dependent on crop ratios and yields which
change, albeit slowly, over time. If so we can easily imagine the economy settling in an equilibrium
to the right of the Malthusian equilibrium. In that equilibrium real wage is constant and above
subsistence, population growth is positive and although the physical unit of land per farmer
continues to decrease the output per farmer is constant. Since technological progress increases
yields and crop ratio we can say that it is ‘land augmenting’ in the sense that the ‘effective’ land is
increasing.
An example
Initially each farmer has 2 units of land, a crop ratio of 0.3 and a yield of 500 kg per unit of land
which means that the yearly output is 2 times 0.3 times 500 which is equal to a yearly output of
300 kg.
After a period of population increase and technological progress affecting the crop ratio and the
yields we have the following configuration
Each farmer has now just 1 unit of land, a crop ratio of 0.5 and a yield of 600 kg, which generates
an output of 300kg.
In the particular case described in the Example box it is assumed that the decline in land per peasant
household caused by population growth is compensated by the combined effect of increased crop
ratio and yield increase leading to a constant real income per head. That is of course an arbitrary
assumption made for explanatory purposes. What happens to output and real income in the long run
will depend on the rate of change of population relative to the rate of change in technological
progress. If the latter is strong income per head might in fact increase, that is output is increasing
faster than population. The long historical trend of increasing population with constant, and in some
areas and periods increasing, income is best explained in the framework discussed here in which
pressure on land is relieved by slow progress in yields and crop ratios.
The demographic transition.
The demographic regime in Europe until the end of the 18th century was characterized by high
fertility and high mortality. Total fertility, that is the expected number of live births per fertile
woman, was very high, on average between four to six births. But given the high mortality rates, in
10
particular for the those under five years of age, the rate of increase of population was not very high,
around 0.1-0.5 per cent per year. Paradoxically, the rate of population growth in the old regime was
not very different from Europe at present. However, total fertility, as defined above, is down to less
than two. That process is accompanied by a sharp rise in life expectancy from 25 to 35 years to 70
to 80 years.
The nature of that transition is depicted in Graph 4.
Graph 4. Old and new total fertility regimes relative to a population growth isoquant of 0.10.5 percent per year.
Childeren
per
woman
(TFR)
6
17th-18th
old regime
0.1- 0.5 percent growth of population
5
Intermediate regime
1-1.5 percent growth
4
Modern regime
0.1- 0.5 percent growth of
population
3
2
1
0
20
30
40
50
60
70
80
Life expectancy at birth
The decline in total fertility starts around mid 19th century in UK and a little later in the rest of
north-western Europe and not until after1900 in most other European nations.
The demographic transition includes a period of rapid population growth caused primarily by a fall
in infant mortality and by an increase in life expectancy of those who survived the difficult first
years. All European nations went through a demographic transition starting with a fall in mortality
followed, after a delay of some 50 years, by a fall in fertility. The whole process took about 150 (+50) years starting as early as the mid or late 18th century in some countries like England and
Sweden and a little later in the rest of Western Europe. However, fertility rates remained high for
several generations before a decline set in. By the mid 20th century both fertility and mortality rates
11
converged to their present low rates, as is demonstrated in Graph 5. It is worth noting that the
demographic transition as described by Graph 5 seems to a be worldwide phenomenon even though
economies like Taiwan, Japan, China and Mexico entered into it much later but also experienced a
much faster transition from the old to the modern regime. Most economies today are in the final
stage of the transition with the exception of a large number of African nations, which are midway
through it, exhibiting high fertility and low mortality, even though mortality is on the rise again due
to the AIDS-epidemic.
Graph 5. Fertility and mortality in the Demographic transition
CBR,CDR per 1000s
50
40
30
20
10
CBR
CD
0
125
The fall in mortality in 20th century developing nations can be explained by better access to
medical services, but that was not the case in mid or late 18th century Europe. Public health
measures did not play an important role until late in the 19th century. In fact the decline in child
mortality is still an unresolved historical event. It is known that an improved nutritional standard of
pregnant women can contribute to higher survival probabilities of the newborn but the fall in child
mortality was not necessarily accompanied by rising income levels. However, given a stable income
level a redistribution of food resources to women might have occurred. It is known that females
generally were less well fed than males either because of outright discrimination and/or because
12
men had physically more demanding work. The hypothesis that women were able to gain access to
a larger share of the household’s food resources is at best unproven. Purely accidental exogenous
factors such as the development of a less aggressive disease environment cannot be ruled out or an
acquired immunity against prevalent diseases. Whether the improvements in temperature, as
indicated by Appendix Graph 1, played a role is an open question.
The other puzzling issue in our understanding of the demographic transition is the fact that fertility
remained high despite falling mortality several generations before it started to decline. You might
expect that household fertility strategy anticipated the high incidence of child mortality. If so the
high total fertility should be interpreted as a necessary and for women in particular, very demanding
task, to bring two or three children out of four to five births to adulthood. If two to three surviving
children were the target size of a family a decline in child mortality should induce families to
reduce the number of births. But that does not happen initially. One possible interpretation is that
the total fertility was in fact the desired outcome in terms of family size which amounts to saying
that the high mortality risks were not anticipated. This is not very convincing but some support for
this interpretation stems from the insight that average mortality risks concealed huge variations
among individual households over time, making it very difficult for households to form correct
expectations of these risks. Another interpretation is that households, given the huge variations in
mortality rates over time, especially in the early phase of the demographic transition, initially
believed the decline to be only transitory. We need to conclude that the early phase of the
demographic transition is still looking for plausible explanations.
The decline in fertility occurs before the widespread use of contraceptives and coincides with the
early phases of urbanization and industrialization and with increasing household income. As
discussed in the Exploratory Box below it is in fact not at all surprising that household size fell
when income increased.
Exploratory Box
Understanding fertility strategies
Malthus, like his contemporaries, did not base fertility strategy on optimizing behaviour of the
household which modern economists and economic historians do. Natural instincts leading to
excessive procreation dominated his thinking even if he also admitted the possibility of moral
restraints on procreation. It is commonly accepted that no known society has fertility rates on the
level of the biological maximum. Also in the absence of modern fertility control methods like the
pill or condoms there were means of controlling fertility, even though these methods were less
13
efficient. For example, postponement of marriage was one fairly efficient way of controlling total
fertility, since births out of wedlock were rare. Coitus interruptus, long periods of breast-feeding
independently and/or taboos on sex during breastfeeding as well as abortions were other methods
of controlling fertility.
An economic approach to the understanding of the fertility strategy of households suggests that
numbers of children cannot alone be what matters but rather the ‘joint product’ of numbers and
quality of children, where quality means education and welfare, such as level of spending on food,
housing, clothing and schooling. Since both numbers and quality are associated with costs there
will be a trade off in the sense that given the income constraint of the household it cannot both
have more and better children. Over time you can expect that households will develop a preference
for quality of children at the expense of quantity because entry into the labour market has become
increasingly demanding in terms of educational attainment.
It is important not just to focus on the income effect* on the demand for the ‘joint product’ but also
the substitution effect*. Since one of the major costs of having children is the income forgone in
giving birth and taking care of children by staying out of the labour market, any income increase
will also automatically increase the opportunity cost* of having children. Modern fertility
strategies in high income nations can therefore be explained by the drift in preferences towards
quality of children as well as strong substitution effects when wages increase, apparently strong
enough to countervail the income effect.
An awkward aspect of the Malthusian population theory is that it postulates that an income
increase is only ‘spent’ on children. However, households must be expected to have preferences for
a wider variety of services and commodities that can be attained by higher income. Income
elasticity* of demand for the ‘joint product’ defined above need not be very high, even though it is
plausibly positive. However, that does not necessarily imply that the number of children demanded
by a household enjoying a rise in income will increase, as explained above.
In the past as well as at present fertility strategies of households will depend on the possibility of
households to externalize the costs of children. For example, if households have access to common
resources the costs will be diffused to the entire community which might lead to fewer restraints on
fertility compared to a context in which households bear the entire cost. It is known, in the past,
that national differences in total fertility might be linked to laws governing the inheritance of
property. If all offspring have equal access to property, as in France after the enactment of Code
Napoleon, family size was reduced compared to nations were only the first born male had legal or
14
customary access to the property. The logic is here supposed to be the desire of households not to
split landed property into small units.
Modern European nations differ in that some, for example the Scandinavian nations, externalize
the costs of raising children by offering subsidized child care financed by taxes paid by parents and
non-parents alike. It is not clear, however, that total fertility in Scandinavia differs significantly
from nations without comprehensive child care at similar level of economic development.
Conclusion
The history of Europe has conventionally been framed in a discussion of the spectre of population
outstripping resources. The approach taken here is to see the slow growth of technological progress
as the major constraint. After all the land constraint is more or less given and unchanged throughout
history. Technology, that is useful knowledge, is not. Agriculture, slowly before 1700, and from
then on at an increasing rate, developed new improved methods increasing crop ratios and yields,
increasing world population from a maximum of 8 million in the pre-neolithic age to a predicted 8
billion in 2025.
.
References:
R.C. Allen, The Nitrogen Hypothesis and the English Agricultural Revolution, A Biological
Analysis. Unpublished paper, Nuffield College, Oxford.
G. Clark, The long march of history: Farm Wages, population and economic growth, England 12091869, The Economic History Review 60, 1, 97-135, 2007
G. Federico, Feeding the World, An Economic History of Agriculture, Princeton: Princeton
University Press, 2005.
M. Kelly, Living Standards and Population Growth: Malthus was Right. Unpublished paper.
University College,Dublin.
R. Lee, The Demographic Transition: Three Centuries of Fundamental Change, Journal of
Economic Literature, 17, 4, 167-190,2003.
M. Livi-Bacci, A Concise History of World Population, 3rd Edition, Oxford:Blackwell,2001.
E.A. Wrigley, et al, English population history from family reconstitution, 1580-1837, (1997)
15
Appendix Graph 1. Real wages in England, northern hemisphere temperature and vital rates for
England 15500
120
Real farm wages (1860-9=100)
100
25 per. Mov. Avg.
(Temperatur)
25 per. Mov. Avg. (CBR)
-0,1
-0,2
80
-0,3
-0,4
60
-0,5
40
-0,6
-0,7
20
-0,8
-0,9
15
4
15 1
51
15
6
15 1
71
15
8
15 1
91
16
0
16 1
11
16
2
16 1
3
16 1
41
16
5
16 1
61
16
7
16 1
8
16 1
91
17
0
17 1
11
17
2
17 1
31
17
4
17 1
51
17
6
17 1
71
17
8
17 1
91
18
0
18 1
11
18
2
18 1
31
18
4
18 1
51
18
61
0
Source: Crude death rates and crude birth rates from E.A. Wrigley,. et al, English population history
from family reconstitution, 1580-1837, (1997). Appendix 3. Vital rates are per thousand.
Real day wage rate for farm workers from G Clark, “The long march of History”, The Economic
History Review, 60 (2007), p. 100. Real wage is indexed 1860-9 = 100.
Temperature is the deviations in Celsius degrees measured on the right hand side of the graph. from
average temperature as of the period 1960-80. Higher temperatures are indicated by a rise in the
curve. From Keith R. Briffa, “Annual climate variability in the Holocene: interpreting the message
of ancient trees”, Quaternary Science Reviews, Volume 19, Issues 1-5, 1 January 2000, Pages 87105
Jan Esper, Edward R. Cook, and Fritz H. Schweingruber “Low-Frequency Signals in Long TreeRing Chronologies for Reconstructing Past Temperature Variability, Science 22 March 2002 295:
Day wage rate: Clark, G., “The long march of History”, The Economic History Review 60 (2007),
p. 100 and 120
16