The Malthusian Model

After reviewing the key development and growth facts, it is clear that we need a theory
that can generate a period of constant living standards, followed by a transition period
with modest increases in the living standard, followed by a period of modern economic
growth. We already have a model that can account for the period of modern economic
growth; the Solow Model with technological change generates constant growth of per
capita output. It is true that the Solow Model can generate a steady state with a constant
level of per capita output. This is the Solow Model absent technological change. One
possibility is to interpret the pre-1700 era of constant livings standards as the steady state
of the Solow Model absent technological change. The problem with this interpretation is
that technology was not stagnant before 1700. Joel Mokyr a noted economic historian at
Northwestern University has documented in his book The Lever of Riches that numerous
and important technological innovations occurred well before 1700.
In light of the historical record on technological change, we proceed to alternative
theory and model of this pre-1700 era. This is the Malthusian model that goes back to
David Ricardo and the classical economists.

There are two key components of the

model. The first is a production function with a fixed factor of production. By fixed, we
mean that its supply cannot be changed over time.

Labor and capital are not fixed

factors as both can be increased over time. In the Malthusian model, the fixed factor is
land. The second key component is a population growth function that is an increasing
function of per capita consumption. These two elements ensure that the steady state is
characterized by a constant living standard even when there is technological change.
We first proceed by studying the Malthusian model with no capital and absent
technological change to help develop intuition for the model. We solve the equilibrium
of the model graphically. We then follow this up with an algebraic study of the
Malthusian model with capital accumulation and exogenous technological change.

Figure (1) below is a plot of per capita income from 3000 BC to 2000 AD. The
feature of the plot is that the standard of living displayed no trend for the first 4800 years
(this period includes Malthusian and Post-Malthusian regimes) and then exploded
subsequent to 1800 (Modern regime). These notes are concerned with a theory of the
period prior to 1800.

Income per Capita

30

20

10

1
0
3000 BC

2000 BC

1000 BC
year

1 AD

1000

2000

Figure 1

I. Model with No Capital or Technological Change

People. Initially, there are N0 people alive. We use Nt to denote the number of people in
the economy at date t. People prefer more consumption to less.

Demographics: Population growth is determined by the death rate and birth rate of the
population. Thomas Malthus proposed a theory of population dynamics in which the birth
rate was independent of peoples living standard.

The death rate however was a

decreasing function of the amount people consumed. This assumption followed from the
idea that if people had more to eat they would be stronger and thus their bodies would be
more able to successfully fight off disease. Graphically, we have
2

Birth and Death Rates as functions

of per capita consumption

death rate

Birth rate

consumption

The population growth rate for a given level of consumption is the difference between the
birth rate and the death rate. We denote this function by g(c). When the birth rate equals
the death rate, the population does not change, namely Nt+1=Nt. More generally,
Nt+1=Nt[1+g(c)].

Population Growth Rate Function (Net)

N t +1
1
Nt

g(c)

0
consumption

In what follows we will use the gross population growth rate function G(c) which is equal
to 1+g(c). Consequently,
N t +1 = N t G (ct ) .
Graphically, we have

Gross Population Growth Rate Function

N t +1
Nt

G(c)

1
consumption

Endowments: Each person in the economy is endowed with one unit of time each period
which he or she can use to work. All individuals own an equal share of the land and the
total amount of land is fixed at L. Land does not depreciate.

Production Function: The economy produces a single final good using labor and land.
The production function is given by

Yt = ALt N t1
The letter A is again the Total Factor Productivity (TFP).

Notice that this production function looks the same as the production function in the
Solow model except that (1-) is labors share and there is land and not capital. The
production function is still characterized by constant returns to scale, and is increasing in
each of its two input. Moreover, the law of diminishing returns applies to each factor
separately. The increases in output associated with an additional unit of labor input
decrease as labor increases holding the land input fixed.
The key feature is that land is essential in production. (Note the last point means that
doubling population while keeping land fixed less than doubles the output).
General Equilibrium:
There are three markets that must clear in each period for the economy to be equilibrium:
the labor market, the land rental market, and the goods market. Equilibrium quantities
are trivially determined in this model, as they were in the Solow Growth model. This is
because the supply of labor and the supply of land are both vertical. People supply their
entire time endowment to the market and their entire land endowment to the market. As
Nt and L are the equilibrium inputs, it is trivial to determine the equilibrium amount of
output. This is Yt = ALt N t1 . For the goods market to clear, Ntct=Yt.

The equilibrium prices are the rental price of land rLt and the real wage rate, wt. To
determine these we need to use the firms labor demand and its demand for land services.
Once again, these are the marginal product of land and the marginal product of labor.
These follow from the profit maximization problem of the firm. This is
Profits: Pr ofits

= A Lt N t1 wt N t rLt Lt

Labor Demand: wt = (1 ) A Lt N t = (1 )

Capital Demand rLt = A Lt 1 N t1 =

Yt
Lt

Yt
Nt

Steady State Analysis

We begin with characterizing the steady state equilibrium. First note that there can not be
any sustained growth in per capita output or consumption in this model. The reason for
this is that the only way a persons output could be increased in this world is by
increasing the amount of land he uses. But land is fixed and so cannot be increased on a
per person basis. Thus, a steady state equilibrium is characterized by a path of per capita
consumption and per capita output that are constant, i.e. zero growth. Once we recognize
this fact, it must be the case that the population is constant in a steady state. Otherwise,
with diminishing returns, per capita output and per capita consumption would have to fall
as we add more people to work the land.
Now, that we recognize this we can be more formal in solving out the steady state
equilibrium. First, we can exploit the result that Nt+1 = Nt = Nss in the steady state with
the population growth rate function to solve for the steady state living standard, css .
Graphically this is just

Finding Steady State Consumption

N t +1
Nt

G(c)

1
css

consumption

Now that we have the steady level of consumption we can solve for the steady state
population of the economy. Again, we this by means of a diagram that consists of two
curves: the hands and the mouth. The mouth curve indicates for any population the
amount of output that is needed to give everyone in the population the steady state
consumption level, css. This is just a straight line from the origin with slope, css. The
hands equation is just the amount of output that is produced with a given population size.
This is just the production function curve. The steady state population is determined by
the intersection of these two curves. This is shown below.

Determining the Steady State Population

Ntcss

Y = AL N 1

N
Nss

Algebraically, what we have done in solving for the steady state is first use the population
growth rate function to solve for css when Nt+1/Nt=1. This is 1 = G (c ss ) . The next step,
which uses the hands and mouth graph, is to solve for Nss using the goods market clearing
condition and the production function equation. Namely, Nc ss = AL N 1 . The left hand
side of the goods market clearing condition is the mouth curve while the right hand side
is the hands.

Comparitive Statics
We can use the diagrams for the population growth function and the hand-mouth to show
how the steady state of the economy is affected by various factors. As the birth and death
rates determine the function G(c), anything that affects either the death rate or the birth

rate will affect the steady state consumption level and population by changing the
position of the mouths curve. Anything that changes the production function will change
the steady state population through its affect of the Hands curve.

The production

function change will not have any effect on the steady state consumption level, however.
Transitional Dynamics
For an economy to be on its steady state it must start out with the right initial population
size, namely, N0=Nss. Suppose and economy fails to start with such a population. Will it
converge to the steady state both in terms of consumption and population? The answer is
yes, it will. This can be seen by using the Hands and Mouth Diagram. Suppose an
economy starts with a population below the steady state level. According to the Hands
equation, output is at Y0. The Mouth equation tells us the amount of output needed to
give each person the steady state consumption level. It is clear that total output exceeds
the required amount. Hence, c0 > css. Now consider what happens to population in
period 1. Here we use the population growth function. As c0 > css, the population growth
function implies that N1/N0 > 1, so population expands. Now at N1, it is still the case that
the Hands output exceeds the Mouth output, so that the living standard, c1>css. However,
c1 is smaller than c0. This can be seen in the Hands and Mouth diagram by noting that the
ray from the origin to any point on the hands curve is equal to per capita output, Y/N,
which is equal to per capita consumption. As N increases, the slope of this ray declines.

10

Population

Consumption

We thus have the following time paths for population and consumption.

The

convergence property is again the result of the law of diminishing returns and the
increasing nature of the population growth function. If population is low, the marginal
product of labor is high. People will therefore

II. Adding Technological change and Capital

Thus far we have shown that the steady state of the Malthusian model is characterized by
a constant living standard and zero population growth. How do these properties change if
we allow for capital accumulation and technological change? Recall, that we are after a
model of the pre-1700 era that generates a constant living standard and population
growth.

11

With the addition of capital it is no longer possible to characterize the properties

of the model graphically. For this reason we will proceed by deriving algebraically the
steady state of the model. It is possible to give some intuition for the algebraic results
that will be derived, however. First, let us think back to the Solow model without
technological change. Was there sustained growth in the per capita consumption and
output in that model without technological change? We showed that sustained growth
was not possible on account of the law of diminishing returns. In the Malthus model the
law of diminishing returns still holds so adding capital to the Malthus model will do
nothing to change the result of a constant living standard. How about the affect of adding
capital accumulation on population growth: Would it change the result of zero population
growth in the steady state?

Again, it would not; you could double all the people and

machines and output would less than double because land is fixed. Here we see the
importance of the fixed factor property of land.
What happens when we add technological change? To gain some intuition here,
let us return to the Malthus model without technological change or capital accumulation
and ask what happens to the steady state following an increase in TFP. This is shown
below. As can be seen, the population is higher. The living standard, however, does not
change since that is determined by the population dynamics of the model. We can think
of technological change therefore, as a sequence of increases in TFP. Consequently, in
the case of technological change we will have a steady state with constant population
growth and a constant living standard- the pre-1700 facts.

12

Increase in TFP

Ntcss

Y = AL N 1

N
Nss

General Equilibrium:
The set of conditions that an equilibrium satisfies are listed below. In addition to the land
and labor market, there is now a capital rental market. Thus, there is its rental price and
the firms demand for capital services given by equation (7). As there is now capital,
there is also savings (as seen in equation 1) and the law of motion for capital (equation 3).
(1) N t ct = (1 s )Yt
(2) Yt = A K t Lt [(1 + m ) t N t ]1
(3) K t +1 = (1 ) K t + sYt
(4) N t +1 = N t G (ct )
(5) wt = (1 )(1 + m ) t (1 ) A K t Lt N t = (1 )Yt / N t
(6) rLt = A K t Lt 1[(1 + m ) t N t ]1 = Yt / Lt

13

(7) rkt = A K t 1 Lt [(1 + m ) t N t ]1 = Yt / K t

Balananced Growth Path Equilibrium. A Balanced Growth path equilibrium is an

equilibrium such that for the right initial population and capital stock endowment, all
variables grow at constant rates, with the possibility that this rate is zero for some
variables.

Notationally, given K0 and N0, ct+1/ct=1+gc, yt+1/yt=1+gy, Yt+1/Yt=1+gY,

Kt+1/Kt=1+gK, Nt+1/Nt= 1+gn, rLt+1/rLt=1+grL, wt+1/wt=1+gw, rkt+1/rke=1+grk.

We divide the solution of the balanced growth path into two parts. In the first, we derive
the growth rates of each of the key variables along the balanced growth path. This is Part
I. In the second part, we actually solve for the paths of the population and the total capital
stock, as well as their initial values.

Part 1. Solving for the growth rates along the balanced growth
path.
Step 1. Use equation (4) to conclude that gc= 0 and ct=css.
Step 2: Use equation (1) to conclude that gn=gY.
Step 3. Use equation (3) to conclude that gK=gY. First divide both sides by Kt. This is
K t +1 / K t = (1 ) + sYt / K t .
Next invoke the steady state condition that Kt+1/Kt=1+gK. This is

1 + g K = (1 ) + s

Yt
Kt

14

As the left hand side of the equation is constant, it follows that Y and K must grow at the
same rate along the balanced growth path.
Step 4. We now use the above results, that g K= gn = gY g with equation (2) to solve for
g. First take the date t+1 output. This is

Yt +1 = A K t+1 Lt +1[(1 + m ) t +1 N t +1 ]1
Next take the date t+1 output as a ratio of the date t output. This is
Yt +1 AK t+1 Lt +1[(1 + m ) t +1 N t +1 ]1
=
Yt
AK t Lt [(1 + m ) t N t ]1
Rearranging terms, we arrive at

K
Yt +1
= (1 + m)1 t +1
Yt
Kt

N t +1

N
t

Now invoke the steady state growth rates

(1 + g ) = (1 + g ) [(1 + m )(1 + g )]1

Lastly, solve for (1+g)
(8)

1 + g = (1 + m , ) (1 ) /

Part II. Solving for the Balanced growth path of K and N.

Step 1. Now that we have solved for the growth rate of the population, we can solve for
css using from the population growth rate function,

Step 2. Use equation (1) to solve for Y as a function of Nt and css. This is
(9)

Yt =

N t c ss
.
1 s

Step 3. Substitute for Yt in equation (3) using equation (9).

15

K t +1 = (1 ) K t +

c ss s
Nt .
1 s

Step 4. Use the BGP condition that Kt+1=(1+g)Kt to solve for Kt. This is

Kt =

(10)

c ss s
Nt
(1 s )( g + )

Step 5. Next take equation (3) substituting for Yt using equation (2). This is
(11) K t +1 = (1 ) K t + sA K t Lt [(1 + m ) t N t ]1 .

Step 6: We again invoke the BGP condition Kt+1=(1+g)Kt and use equation (10) in (11).

sc ss
( g + ) = sA

(1 s )( g + )

N t 1 Lt [(1 + m ) t N t ]1

This gives us a single equation in a single unkown, Nt. Solving for Nt yields

(12)

ss
t

= L(1 + m )

t (1 )

sA

g +

sc ss

(1 s )( g + )

1 1 /

Using equation (12) with (10), gives us the solution for K tss

III. Computing the Equilibrium path when

N 0 N 0ss or K 0 K 0ss .

Convergence
If N 0 N 0ss or K 0 K 0ss , the economy will not be on its balanced growth path. We
showed graphically for the Malthusian model without capital accumulation and
technological change that an economy which does not begin on its steady will converge
to its steady state. Although we cannot show it graphically for this more complex version
of the model, the convergence result holds.
16

It is easy to compute the transitional path for the model economy via a
spreadsheet once we assign parameter values to the model. This is discussed in the last
section.

The equilibrium consists of a path for the following set of variables:

(ct , N t , Yt , Lt , K t , wt , rLt , rKt ) .

The equilibrium quantities are trivial to find because the supply of land, labor and
capital are all vertical. That is to say in period t, the amount of capital used is Kt; the
amount of labor used is the population, Nt; and the amount of land used is just the
endowment L. We therefore know what Yt and ct are. We also know what the margina
products of all the inputs are, so we know all of the prices. We can then figure out next
periods population and capital stock using the population growth rate function and the
capital stock law of motion. The period t=0 quantities of Land, capital and people are
given.
More specifically, we compute the equilibrium path as follows. , K0 , and an
initial population N0, we find the equilibrium path in each period as follows:
1. Determine Yt from the aggregate production function, Equation(2).
2. Determine ct from the

IV. Calibration to pre-1700 Observations

We can calibrate the model so that it matches the pre-1700 growth facts.

We need to

assign values to the following list of parameters: , , m, , s, A . In addition we need to

specify the population growth function and assign parameter values to that function. In
what follows, we assign parameter values so that the model matches the following pre1700 observations associated with Englands experience.

17

1. an average annual population growth rate of .003 per year.

2. Historians estimates of labors share of income equal to 2/3.
3. Historians estimates of Lands share of income at 3/12.
4. Historians estimates of capitals share of income at 1/12.
We continue to set the savings rates, s= .20 that we used for the Solow model.
Additionally, we continue to use the depreciation rate that we used for the Solow model,
namely, = .05. Just as in the Solow model, an inputs share of income equals its
coefficient in the production function. For land, this coefficient is ; for capital, this
coefficient is ; and for labor this coefficient is just 1 . Thus, = 3 / 12 and

= 1 / 12 . We can determine the value of m using the balanced growth relation between
population growth and the exogenous growth rate of technological change. This is
1 + g = (1 + m , ) (1 ) / .
Using the observation that g=.003 and the values for = 3 / 12 and = 1 / 12 , we can
solve for the value of m. This is m = .001 . The last technology parameter, A, is just
normalized for one as it determines the units in which output is measured.

The

parameterized production function is thus

Yt = AKt1 / 12 L3t / 12 [(1.001)t N t ]2 / 3 .

The Population growth function
For the predictions of Malthus to hold, we need a population growth function that is
increasing in the living standards that governed the pre-1700 era. As the model predicts a
constant population growth along the balanced growth path, it is clear that we need an
episode where the English economy was not on its balanced growth path. Such an
episode is the Black Death that occurred between 1347 and 1350. The effect of the Black

18

Death on Europes population was dramatic; historians estimate that one half to one third
of Europes population died in this three year period. While deaths were concentrated in
this three year period, it took nearly 200 years for the population to return to its pre 1346
level in Europe.
Our strategy is thus to use the data on living standards (actually real wages) and
population growth for available years in the 1350 to 1550 period to estimate a linear
equation for population growth. Namely,

N t +1 / N t = 0 + 1ct .

0 is the y-intercept of this linear relation and 1 is the slope. The observations for the
English economy for living standards and population growth are shown in the following
figure:

19

Population Growth Rate for England between 1350 and 11550

1.015

1.01

1.005

0.995

0.99
80

90

100

110

120

130

140

150

160

170

real wage

The estimated slope line is also depicted. This line is Nt +1 / N t = .999 + .00004ct . This is
the function G(c).

V. Empirical Support for the Malthusian Theory

We conclude this chapter by presenting some evidence in support of the Malthusian
theory.

The English Economy From 1250 to the Present

A. The Period 12751800
The behavior of the English economy from the second half of the 13th century
until nearly 1800 is described well by the Malthusian model. Real wages and, more
generally, the standard of living display little or no trend. This is illustrated in Figure 1,

20

which shows the real farm wage and population for the period 12751800.i During this
period, there was a large exogenous shock, the Black Death, which reduced the
population significantly below trend for an extended period of time.

This dip in

population, which bottoms out sometime during the century surrounding 1500, is
accompanied by an increase in the real wage. Once population begins to recover, the real
wage falls. This observation is in conformity with the Malthusian theory, which predicts
that a drop in the population due to factors such as plague will result in a high labor
marginal product, and therefore real wage, until the population recovers.
Population and Real Farm Wage
300
250
200

Wage

150
100
Population

50
0
1275

1350

1425

1500

1575

1650

1725

1800

Figure 1
Another prediction of Malthusian theory is that land rents rise and fall with
population. Figure 2 plots real land rents and population for England over the same
12751800 period as in Figure 1.ii Consistent with the theory, when population was
falling in the first half of the sample, land rents fell. When population increased, land
rents also increased until near the end of the sample when the industrial revolution had
already begun.

21

Population and Real Land Rent

300

250

200

150
Population

100

Rent

50

0
1275

1350

1425

1500

1575

1650

1725

1800

Figure 2

The English population series is from Clark (1998a) for 12651535 (data from parish records in 1405-

1535 is unavailable, so we use Clarks estimate that population remained roughly constant during this
period) and from Wrigley et al. (1997) for 15451800. The nominal farm wage series is from Clark
(1998b), and the price index used to construct the real wage series is from Phelps-Brown and Hopkins
(1956). We have chosen units for the population and real wage data so that two series can be shown on the
same plot.
ii

The English population series and the price index used to construct the real land rent series are the same

as in Figure 1. The nominal land rent series is from Clark (1998a).

22