The Solow Model
Lecture 1
Econ 314
Particulars of Solow (1956)
Accepts all the Harrod-Domar assumptions except fixed
proportions.
Harrod-Domar speak of the long run in terms of the multiplier,
the accelerator and the capital coefficient.
One thinks of the long run as the domain of the neo-classical
analysis, the land of the margin.
Harrod-Domar studies the long run with short run tools. Solow
departs from this point onwards.
The original formulation of the Solow Model
• At every instant, output, Y(t) is consumed, saved and invested.
• Rate of savings = sY(t)
• Stock of capital = K(t)
• Net investment = rate of increase in capital stock = dK/dt = K̇
K̇ = sY(t) (1)
Consider output to be net output after compensating for depreciation of
capital. Then,
Y = F(K,L) (2)
Production at each instant shows CRS. So the production function is
homogeneous of degree one. Inserting (2) in (1),
K̇ = s F(K,L) (3)
The original formulation of Solow Model
• There are 2 unknowns in equation (3). It can be closed with an equation each
for the demand and supply labor.
• However, Solow defines the supply of labor as perfectly inelastic and assumes
full employment.
• Assuming exogenous population growth rate ‘n’
L(t) = Lo ent (4)
In (3) L = total employment. In (4) L = supply of labor.
Assuming full-employment, and inserting (4) in (3), we have
K̇ = s F(K, Lo ent) (5)
The original formulation of Solow Model
• Solution of (5) gives time profile of capital stock which will fully
employ the available labor.
• Knowing the time profile of capital stock and labor force, the time
path of real output can be calculated from production function.
• To check whether there is always a capital accumulation path
consistent with any growth rate of labor force, we study (5).
However, we need to specify the shape of the production function.
Thus, introduce k = K/L
Therefore, K = kL = kLo ent . Differentiating wrt. time,
K̇ = k̇ Lo ent + nk Lo ent
The original formulation of Solow Model
Substituting in (5),
Lo ent [k̇ + nk] = s F(K, Lo ent)
Due to CRS, we can divide both variables in F(K, Lo ent) by L= Lo ent
provided we multiply F(K, Lo ent) by the same factor.
Lo ent [k̇ + nk] = s Lo ent F(K/ Lo ent, 1)
Which reduces to,
k̇ = s F(k, 1) - nk (6)
The original formulation of Solow Model
k = K/L which implies, k̇ /k = K̇ /K – Ĺ/L ; Ĺ/L = n
When k̇ = 0 , K/L is constant, K̇ /K = Ĺ/L = n
And nk = s F(k, 1) ; k̇ = k*
When k > k*, nk > s F(k, 1) ; k → k*
k < k*, nk < s F(k, 1) ; k → k*
Thus, k* is a stable equilibrium.
The time path of capital and output will not be exactly exponential
except asymptotically.
If k < k*, capital & output grows faster than labor, and,
If k > k*, capital & output grows slower than labor, till k* is approached.
Note that we shall find this in the equation of motion of capital in our own presentation of
the Solow model hereafter.
A Classroom Presentation of the
Solow Model
Assumptions
Labor and capital are the two factors of production
The level of technology is exogenously given
There are constant returns to scale
The production function is a Cobb-Douglas
production function.
Constant returns to scale
If the factors of production are increased by a given number, the output
increases by the same number as well. Thus, if the production function
is defined as
Y = F(K,L)
Then, multiplying K and L by a factor ‘t’, F(tK, tL) implies
F(tK, tL) = tF(K,L) = tY
Let Y = F(K,L) = KαL(1- α) ; Let ‘t’ = 2
Then multiplying K and L by 2, we have F(tK, tL) = [2K] α[2L](1- α)
= 2 α+(1- α)KαL(1- α)
= 2KαL(1- α)
= 2F(K,L)
= 2Y
Thus, Y = KαL(1- α) is a production function, with CRS.
Cobb-Douglas production function and CRS
Y = AKα Lβ is a Cobb-Douglas production function
CRS requires, α + β = 1
Thus, the Cobb-Douglas production function used in
the Solow model, becomes Y = AKα L(1-α)
To reduce the above to a per capita production,
Y/L = A[K/L]α [L/L](1-α)
→ y = Akα
The Production Curve
Output per capita
y = f(k) f(k)
The production curve obeys the
law of diminishing marginal
productivity of capital
(0,0)
Capital per capita k
y = f(k) The Steady State δk
y* f(k)
The Steady State level
of output per capita
sf(k)
The Steady State level
of capital per capita
(0,0)
k* k
The Steady State – a stable equilibrium
y = f(k) (δ+n)k
y* f(k)
Excess
Investment
sf(k)
Excess
Savings
(0,0)
k1 k* k2 k
The impact of lowering the savings rate
y = f(k) (δ+n)k
y* f(k)
y**
sf(k)
s’f(k)
s’ < s
The steady state level of
capital per capita declines
(0,0)
k** k* k
Growth of per capita variables at the steady state
sy = sf(k) = skα represents savings per capita
(n + δ)k, represents, the amount of investment needed to keep per capita
capital constant. Here, n = rate of growth of population ; δ = rate of
depreciation.
The level of per capita capital at which
k(dot) = skα − (n + δ)k = 0,
is the “steady-state capital stock per capita”, and denoted k*.
At the steady state therefore, the change in the per capita variables,
k(dot) = 0 and y(dot) = 0
Accordingly, the rate of growth of the per capita variables, [k(dot)/k] and
[y(dot)/y] are zero as well.
Level of per capita variables at the steady state
In the steady-state k(dot) = 0.
Therefore,
sk*α = (n + δ)k*
k*(α−1) = (n + δ) / s
k* = [s / (n + δ)] 1/(1-α)
Per capita income then satisfies:
y* = k*α = [s / (n + δ)]α/(1- α)
The steady state level of income per capita is:
1. Increasing in the savings rate
2. Decreasing in the population growth rate
3. Decreasing in the deprecation rate
Growth of aggregate variables at the steady state
In the steady-state, aggregate capital, K(t), is not constant:
K(dot) / K = [k(dot) / k] + [L (dot) / L] = n > 0
Similarly, total output, Y (t) grows as well:
Y = KαL(1−α)
ln Y = α ln K + (1 − α) ln L
So,
Y(dot) / Y = α [K(dot) / K] + (1 − α) [L(dot) / L]
= αn + (1 − α)n
= n.
Therefore, the aggregate capital and output, each grow at a rate equal to
the rate of growth of population, “n”
Conclusions and Criticisms
The per capita variables grow at a rate ‘zero’ while the aggregate
variables grow at a constant rate of growth of population.
A change in the rate of savings or depreciation can change the
rate of growth temporarily till the economy reaches a new steady
state, where the rate of growth of the per capita variables are
‘zero’ once more.
A change in the rate of growth can only be brought about by a
change in the level of technology, which is exogenously given in
the model.
The economy described in the model is a closed economy, so
that the growth potentials from foreign aids and loans have not
been looked into.
