lOMoARcPSD|1200466

Solow Model of Economic growth

Economics (Panjab University)

StuDocu is not sponsored or endorsed by any college or university

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)
 lOMoARcPSD|1200466

Solow Model of Economic

Growth

Work of:-

Manhar Manchanda
Saksham Diwan

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

BRIEF CONTENTS

 Introduction 4
 A Model of Long Run Growth 4
 The Solow Equations 6
 Diagrammatic Representation: The Steady State 7
 Possible Growth Patterns 9
 Technology and the Solow Model 12
 Solow Diagram with Technology 13
 Convergence 14
 Augmented Solow Model- A New Approach 22

 References 26

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

“All theory depends on assumptions which are not quite true. That is what makes it theory.”
-Robert M. Solow (1956)

I. INTRODUCTION

Being aware of the fallacies that constitute the unrealistic nature of the Harrod-Domar
growth model, Robert M. Solow, Nobel Laureate in Economics (1987), devised a model with
the suitable tools to tackle the problems that emerge in an economy in the long run.
“The characteristic and powerful conclusion of the Harrod-Domar line of thought is that
even for the long run the economic system is at best balanced on a knife-edge of
equilibrium growth.” (Solow, 1956)
According to Solow, the fundamental opposition of warranted and natural rates of growth
turns out in the end to flow from the crucial assumption that production takes place under
conditions of fixed proportions. There is no possibility of substituting labor for capital in
production, and if this assumption was to be abandoned, the knife edge notion of unstable
balance seems to dissipate with it.
The Solow Growth Model is a model of capital accumulation in a pure production economy
(an economy with no exchange i.e. a closed economy) and is simply interested in only
output=real income hence there is no scope for prices. Though the Solow model doesn’t
always have realistic assumptions, it’s assumed that people work all the time, so there is no
labor/leisure choice and they save, hence invest, a fixed portion of their income. In essence,
we assume that people have no choice at all. (Retrieved from-
https://www.unc.edu/~jbhill/Solow-Growth-Model.pdf)

Solow’s growth model hereby follows the footsteps drawn by the Harrod-Domar model,
accepting every assumption put forth in the model barring the assumption of fixed
proportions. Instead, he was of the opinion that under standard neoclassical conditions,
labor and capital are capable of producing a single composite commodity (relative prices
remain unchanged).
“Although in some respects Solow’s model describes a developed economy better than a
developing one, it remains a basic reference point for the literature on growth and
development. It implies that economies will conditionally converge to the same level of
income if they have the same rates of savings, depreciation, labor force growth, and
productivity growth. Thus, the Solow model is the basic framework for the study of
convergence across countries.” (Todaro and Smith, 2015, p.155)

II. A MODEL OF LONG RUN GROWTH

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

The homogeneity assumption is of a rather broad horizon and can be delineated better
using quantitative modeling of the theory. There is only one commodity, output as a whole,
whose rate of production is designated Y(t). Part of each instant’s output is consumed and
rest is saved and invested. The fraction of output saved is a constant s, so that the rate of
saving is sY(t). The community’s stock of capital K(t) takes the form of an accumulation of
the composite commodity. Net investment is the rate of increase of this capital stock dK/dt,
so we have the basic identity at every instant of time:

(1) = sY

Technological possibilities are represented by the production function:

(2) Y = F(K,L)
Inserting (2) in (1), we get:

(3) = sF(K,L)

This is one equation in two unknowns. One way to close the system would be to add a
demand for labor equation: marginal physical productivity equals real wage rate (w); and a
supply of labor equation. In any case, there would be three equations in the three
unknowns K, L, and w.
As a result of exogenous population growth the labor force increases at a constant relative
n. In the absence of technological change n is Harrod’s natural rate of growth. Thus:
(4) L(t) = L0ent
In equation (3), L stands for total employment whereas in equation (4), L stands for the
available supply of labor. By identifying the two, we are assuming that full employment is
perpetually maintained. Inserting (4) in (3), we get:

(5) = sF(K,L0ent)

Equation (5) determines the time path of capital accumulation that must be followed if all
available labor is to be employed.
The whole process can be viewed in this way: at any moment of time, the available labor
supply is given by (4), and the available stock of capital is also a datum. Since, the real return
to factors will adjust to bring about full employment of labor and capital, we can use the
production function (2) to find the current rate of output. (Solow, 1956)

III. (a) The Solow Equations

Solow’s twist on the Harrod–Domar story is based on the law of diminishing returns to
individual factors of production. According to the Solow thesis, the capital–output ratio ‘θ’ is
endogenous. To understand the implications of this modification, it will help to go through a

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

set of derivations very similar to those we used for the Harrod–Domar model. (Debraj Ray,
1998, p.66)
Retaining the assumptions of Harrod-Domar model:
(I) S(t) = I(t)
(II) K(t+1) = (1-d)K(t) + I(t) ; where a fraction ‘d’ of capital stock depreciates
Combine (I) and (II) to get:

(III) (1+n)k(t+1) = (1-d)k(t) + sy(t), where lowercase k’s and y’s represent per capita
magnitudes (K/P and Y/P respectively).

The right-hand side has two parts, depreciated per capita capital [which is (1 – d)k(t)] and
current per capita savings [which is sy(t)]. Added together, this should give us the new per
capita capital stock k(t + 1). The left-hand side of (III) has the rate of growth of population
(n) in it because of the downward drag on per capita capital due to growing population.
Note that the larger the rate of population growth, the lower is per capita capital stock in
the next period.

III. (b) Diagrammatic Representation: The Steady State

Figure 3.1 (source: Debraj Ray, 1998)

Figure 3.1 shows a typical production function with diminishing returns to per capita capital,
i.e., output-capital ratio has a negative relation with per capita capital (because of a relative
shortage of labor). With constant returns to scale, we may use the production function to
relate per capita output to per capita input.

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Figure 3.2 (a) Figure 3.2(b)

(Source: Debraj Ray, 1998)

The production function is used to determine the value of per capita capital stock at time
(t+1); given that current per capita stock is k.
Given above is the diagrammatic representation of the (III) equation that has been derived
in the previous section (in both (a) and (b)). The resulting curve resembles the production
function curve. The straight line (1+n)k is upward sloping due to constant returns to scale
and is plotted to reflect the changes in k. Because of the diminishing returns, the curved line
initially lies above the straight line and then falls below.
In figure 3.2(a), the output-capital ratio is really high accruing to the low stock of capital
hence a per capita capital stock can expand rapidly. The expansion of per capita capital
continues till point k*, which is a distinguished capital stock level where the curved and
straight lines interact. Solow model loses its momentum, to the left of k*, when capital is
growing faster than labour.
In figure 3.2(b), there exists a high initial capital stock. Hence, as time passes, there is an
erosion of the per capita stock (population growth outstrips the rate of growth of capital)
with convergence occurring back to the per capita stock k*. In other words, in this case, the
output-capital ratio is low which further erodes the expansion of aggregate capital.
Therefore, k* is considered to be a steady state level of the per capita capital stock, to which
the per capita capital stock at any initial level must converge. Per capita capital stock settling
down to some ‘steady state’ implies that per capita income must follow.
Thus, in Solow’s Growth model, there is no long-run growth of per capita output, and total
output grows precisely at the rate of growth of the population. In particular, the savings rate
has no long-run effect on the rate of growth, in sharp contrast to the prediction of the
Harrod–Domar model. (Debraj Ray, 1998, p.68)

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

The ‘diminishing returns to capital’ feature, which creates endogenous changes in the
capital-output ratio, is the factor which chokes off growth in the Solow Model. This opposes
the Harrod-Domar model wherein a steady state level (k*) is implausible.

IV. POSSIBLE GROWTH PATTERNS

To see if there is always a capital accumulation path consistent with any rate of growth of
the labour force, we must study the differential equation (5) for the qualitative nature of its
solutions. Naturally, without specifying the exact shape of the production function we can't
hope to find the exact solution. (Solow, 1956, p.5)

(5) = sF(K,L0ent)

We introduce a new variable r=K/L; the capital-labor ratio.

K = rL = rL0ent
Differentiating the above equation w.r.t time, we get

= L0ent + nrL0ent

Substituting in (5)

( + nr)L0ent = sF(K,L0ent)

Dividing and multiplying variables in F by L 0ent due to constant returns to scale, we get:

( + nr)L0ent= sFL0ent( , 1)

(6) = sF(r,1) – nr ; a differential equation involving isolated capital-labor ratio

The equation (6) can be interpreted as the rate of change in capital-labor ratio being the
difference of two terms- one representing the change of capital, and the other representing
the change of labor.
Case-1: When dr/dt=0, i.e., the capital-labor ratio is constant at rate n. This signifies that the
warranted rate of growth (warranted by the appropriate real rate of return to capital)
equals the natural rate.

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

FIGURE: G1
(Source: Solow, 1956)
This is the same case as summarized in the Solow model mentioned in the aforementioned
section, wherein steady state level is achieved at r*. Whatever the initial value of the
capital-labor ratio, the system will develop toward a state of balanced growth at the natural
rate.
Solow has pointed to many other configurations which are a priori possible.

Case-2: Three Intersection Points

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

FIGURE: G2
(Source: Solow, 1956)
The initially derived capital-output ratio allows the system to develop balanced growth at
capital-labor ratio r1 and r3.
In either case labor supply, capital stock and real output will asymptotically expand at rate
n, but around r1 there is less capital than around r3, hence the level of output per head will
be lower in the former case than in the latter. The relevant balanced growth equilibrium is
at r1 for an initial ratio anywhere between 0 and r2, it is at r3 for any initial ratio greater
than r2. The ratio r2 is itself an equilibrium growth ratio, but an unstable one; any accidental
disturbance will be magnified over time. (Solow, 1956, p.8)
Case-3: No steady state level is possible

FIGURE:G3
(source: Solow, 1956)

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

In figure G3, two possibilities are shown, both having diminishing marginal productivity
throughout. One lies wholly above nr while the other lies wholly below. In other words, the
first system is highly productive and hence capital and income both increase more rapidly
than the labor supply. The opposite second system is so unproductive that full employment
path leads only to forever diminishing income per capita.
‘’The basic conclusion of this analysis is that, when production takes place under the usual
neoclassical conditions of variable proportions and constant returns to scale, no simple
opposition between natural and warranted rates of growth is possible.’’ (Solow, 1956)

V. TECHNOLOGY AND THE SOLOW MODEL

To generate sustained growth in per capita income, we introduce technology variable, A, to

the production function (2):

(a) Y = F(K,AL) = Kɑ(AL)1-ɑ

A is said to be ‘’labour augmenting’’ or “Harrod-neutral”. Technological progress occurs
when A increases over time—a unit of labour, for example, is more productive when the
level of technology is higher. We make the assumption that A is growing at a constant rate:
A=A0egt ; where g is a parameter representing the growth rate of technology

Rewriting the capital accumulation equation: = sY-dK (wherein sY is the amount of grosss
investment and dK is the depreciation that occurs during the production process):
( / )
(b) = s( ) - d

Writing (a) in terms of output per worker,

(c) y = kɑ(A)1-ɑ
Taking logs:
logy = (ɑ)logk + (1-ɑ)logA
Differentiate w.r.t. time:
( / ) ( / ) ( / )
(d) =ɑ + (1-ɑ)

We can infer from the capital accumulation equation, (b), that the growth rate of capital K
will be constant if and only if Y/K is constant which further gives us that y/k will also be
constant and ensuring that y and k grow at the same rate. A situation in which capital,
output, consumption, and population are growing at constant rates is called a balanced
growth path.
gy=gk=g gives us the relationship for a balanced growth path wherein, g= (𝑑𝐴/𝑑𝑡)/𝐴 and gy
and gk reflect that output per worker and capital per worker both grow at the rate of

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

exogenous technological change which is nothing but g. The model with technology reveals
that technological progress is the source of sustained per capita growth.

VI. THE SOLOW DIAGRAM WITH TECHONOLOGY

Relaxing the assumption of constant k in the long run, we take a new state variable

(e) k˜= ; which represents the ratio of capital per worker to technology

We refer to this as capital-technology ratio (keeping in mind that the numerator is capital
per worker rather than the total level of capital). This is equivalent to 𝑘/𝐴 and is obviously
constant along the balanced growth path because gK= gA=g.
Writing (c) i.e. the production function in terms of k, we get:

y˜=k˜ɑ ; where y˜= = , is the output-technology ratio

Taking log in (e):

log( k˜) = log(K) - log(A) - log(L)
Differentiating w.r.t. time:
( ˜/ ) ( / ) ( / ) ( /
= – –
˜

Combining this with capital accumulation equation (b):

˜
(f) = sy˜ - (n+g+d)k˜

This equation says that the change in capital per worker each period is determined by three
terms. The equation shows change in capital because of n, population growth, g,
technological growth and d which is depreciation.
Solving for the steady state, we use the equation (f) taking the condition that (dk˜/dt)=0,
which gives us:
0 = sy˜- (n+g+d)k˜
˜
k˜=

We know that y˜= k˜ɑ

˜
k˜=

k˜k˜-ɑ =
( )

Therefore,

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

/ ɑ
(g) k˜* = [ ]
( )

Solow diagram with technological progress

(source: Jones, 2013)
In the above figure, if the economy begins with a capital-technology ratio that is below its
steady-state level, say at a point such as k0˜, the capital technology ratio will rise gradually
over time. This is because the amount of investment being undertaken exceeds the amount
needed to keep the capital-technology ratio constant. This will be true until sy˜= (n + g + d)k˜
at the point k˜* , at which point the economy is in steady state and grows along a balanced
growth path.
Now, substituting (g) into the production function yields
/ ɑ
y˜* = [ ]
( )

To see what this implies about output per worker, rewrite the above equation as
/ ɑ
(h) y*(t) = A(t) [ ]
( )

From equation (h), we see that output per worker along the balanced growth path is
determined by technology, the investment rate, and the population growth rate.

VII. Convergence

7.1 Unconditional convergence

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

The Solow model postulates that countries, in the long run, display no tendency to display
differences in the rates of technological progress, population growth, savings, and capital
depreciation. This ensures that the capital per efficiency unit of labour converges to the
common value k˜*, irrespective of the initial state of these economies (starting levels of per
capita income).

Fig 7.1 Illustration of Unconditional convergence

In the above figure, log of per capita income against time is plotted to show constant rate of
growth as a straight line. The line AB represents time path of log per capita income at steady
state. The line CD represents an economy which starts below the steady state level, whereas
the line EF is of a nation starting above the steady state.
In the case of CD, the time path of log per capita income moves asymptotically toward the
line AB, i.e., over time, growth rate will decelerate to the steady state level. Likewise, time
path EF of log per capita income flattens out to converge to the line AB from above,
experiencing a lower rate of growth.
Convergence, therefore, is indicated by a strong negative relationship between growth rates
of per capita income and the initial value of per capita income.

7.2 Empirical Evidence of Unconditional Convergence

As per the Solow model, unconditional convergence not only requires convergence of
countries to their own steady states but also assumes that all the steady states are the
same.

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

“An early hypothesis proposed by economic historians such as Aleksander Gerschenkron

(1952) and Moses Abramovitz (1986) was that, at least under certain circumstances,
“backward” countries would tend to grow faster than rich countries, in order to close the
gap between the two groups.” (Jones, 2013, p.63)
William Baumol (1986) was one of the first economists to provide statistical evidence of
convergence among some countries and absence among others.

Figure 7.2(a)

The above figure plots per capita GDP (on a log scale) for several industrialized economies
from 1870 to 2008. The narrowing of the gaps, convergence, between countries is evident in
the figure.

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Figure 7.2(b)

Figure 7.2(b) displays the ability of convergence hypothesis to explain why the growth rate
varies from country to country. The graph plots a country’s initial per capita GDP (in 1885)
against the country’s growth rate from 1870 to 2008, exhibiting a negative relationship
between the two variables.
“Countries such as Australia and the United Kingdom, which were relatively rich in 1870,
grew most slowly, while countries like Japan that were relatively poor grew most rapidly.
The simple convergence hypothesis seems to do a good job of explaining differences in
growth rates, at least among this sample of industrialized economies.” (Jones, 2013, p.68)

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Figure 7.2(c)

Figure 7.2(c) examines the convergence hypothesis in the case of Organization for Economic
Cooperation and Development (OECD), for the period 1960-2008. The convergence
hypothesis holds well for explaining varying growth rates across the OECD. However, new
members like Chile and Mexico display growth rates lower than the expected rate.

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Figure 7.2(d)
Figure 7.2(d) plots growth rates versus initial GDP per worker for countries that are
members of OECD and the world for the period 1960-2008. ‘(However, the figure) shows
that the convergence hypothesis fails to explain differences in growth rates across the world
as a whole. Baumol (1986) also reported this finding: across large samples of countries, it
does not appear that poor countries grow faster than rich countries. The poor countries are
not “closing the gap” that exists in per capita incomes.’ (Jones, 2013, p.66)
This empirical failure can be accrued to the assumption for all countries to have similar
technology levels, investment rates, and population growth; hence, they are expected to
grow toward a similar steady state level.

7.3 Conditional Convergence

The principle of transition dynamics states that the further an economy is “below” its steady
state, the faster the economy should grow. The further an economy is “above” its steady
state, the slower the economy should grow. Mankiw, Romer, and Weil (1992) and Barro and
Sala-i-Martin (1992) have termed this concept as “conditional convergence”.
Conditional convergence reflects differences in the steady states of countries, hence they
exhibit a varying convergence. Therefore, it does not mean that all countries are converging

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

to the same steady state, only that they converge to their own steady states according to a
common theoretical model.
The conditional convergence phenomenon says that the countries that have not reached
the steady state are not expected to grow at the same rate. Countries may not be in a
steady state due to various reasons. Economically turbulent activities such as a dramatic
increase in the investment rate, a change in the population growth, or an event like World
War II, generate a gap between current income and steady state income. This gap causes
temporary changes in growth rates until the economy is restored at the steady-state level.

7.4 Evolution of Income distribution

Convergence can have several possible outcomes; the closing of the gap between rich and
poor economies, or the poorest countries falling behind while countries with “intermediate”
incomes are converging toward the rich, or countries are completely diverging, i.e., rich
countries are getting richer and the poor countries are getting poorer.

Figure 7.4(a)

Figure 7.4(a) shows that the enormous gap in income distribution across countries has not
narrowed down with time. This figure plots the ratio of GDP per worker for the country at
the 90th percentile of the world distribution to the country at the 10th percentile.

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

In 1960, GDP per worker in the country at the 90th percentile was about twenty times that
of the country at the 10th percentile. By 2000 this ratio had risen to forty, and after jumping
to about forty-five for a few years, it has returned to around forty in 2008. (Jones, 2013)
Over time, it can be concluded, the gap between the incomes of the richest countries and
the poorest countries rises. Lant Prichett (1997), in his paper ‘Divergence: Big Time’
calculated that the ratio of per capita GDP between the richest and poorest countries in the
world was only 8.7 in 1870 but rose to 45.2 by 1990. One possible explanation for the rising
income distribution gap is that countries start the process of economic growth at different
points in time.

Figure 7.4(b)

Figure 7.4(b) demonstrates the evolution of the world income distribution, i.e., it shows the
percentage of world population at each level of GDP per worker (relative to USA).
According to this figure, “in 1960 about 60 per cent of the world population had GDP per
worker less than 10 per cent of the U.S. level. By 2008 the fraction of population this far
below the United States was only about 20 per cent. This fall can be attributed, in large part,
to increased GDP per worker in China and India. Overall, in both years about 80 per cent of
the world’s population had GDP per worker of less than 50 per cent of the U.S. level.”
(Jones, 2013, p.74)

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Sala-i-Martin (2006) documents the number of people living below different poverty levels
over time. In 1970, 534 million people—about 15 per cent of the world population—lived on
less than $1 per day (in 1996 dollars), the World Bank’s official poverty line. By 2000 only
321 million were at this low level of income, and they accounted for only 6 per cent of world
population. Absolute poverty has been decreasing over time for the world population. In
relative terms, the poorest countries in the world are poorer than they were about fifty
years ago, suggesting that there is a divergence across countries over time. (Jones, 2013,
p.75)

VIII. AUGMENTED SOLOW MODEL- A NEW APPROACH

VIII.1 Theory
Gregory Mankiw, David Romer, and David Weil (1992), in their seminal work “A Contribution
to the Empirics of Economic Growth” noted that the “fit” of the model could be improved by
taking into consideration human capital, i.e., by recognizing that labour in different
economies may possess different levels of education and different skills.
Suppose that output, Y, in an economy is produced by combining physical capital, K, with
skilled labour, H, according to a constant-returns, Cobb-Douglas production function
(8.1) Y = Kɑ(AH)1-ɑ ; where A represents labour-augmenting technology that grows
exogenously at rate g.
Taking the assumption that u, unskilled labour’s time spent on learning new skills generates
skilled labour H, i.e.
(8.2) H=eΨuL ; where L represents the total amount of raw labour used in
production in the economy, and ψ is a positive constant.
If u=0, then H=L, which gives us that all labour is unskilled. Therefore, an increase in u
increases the effective units of skilled labour H.
Taking log of equation (8.2)
logH= ψ u + logL
( / )
=ψ or

(8.3) = ψH

To understand this equation, suppose that u increases by 1 unit (say, 1 additional year of
schooling), and suppose ψ = 0.10, H rises by 10 per cent. To summarize, there exists a direct
proportional relationship between ψ and H, i.e., it intends to match a large literature of labour
economics which gives that an additional unit of learning gives an increase to the wages
earned.

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Now, we rewrite the production function (c) mentioned in Section V, introducing the human
capital, in terms of output per worker, i.e.
(8.4) y=kɑ(Ah)1-ɑ
Here, h = eΨu, we assume that u is constant and determined exogenously, which means that h
is constant. Along a balanced growth path, y and k will grow at a constant rate g.
When introducing human capital in the concept of technological change given by the
extended Solow Model, we get a similar equation derived in Section VI which summarizes
the explanation of why some countries are rich while others are poor.
‘Countries are rich because they have high investment rates in physical capital, spend a large
fraction of time accumulating skills (h=eΨu), have low population growth rates, and have high
levels of technology. Furthermore, in the steady state, per capita output grows at the rate of
technological progress, g, just as in the original Solow model.’ (Jones, 2003, p.57)
We can now introduce human capital in equation (h) of section VI.
/ ɑ
(8.5) y*(t) = hA(t) [ ]
( )

This equation shows the steady state value of output-technology ratio y˜ in terms of output
per worker along the balanced growth path (y*(t)).

VIII.2 Empirical Analysis

We analyse the model in terms of relative incomes because incomes are growing over time.
Defining per capita income relative to the United States:
y^*(t) = y*/y*US
Then from equation (8.5), we get
(8.6) y^*= h^A^(s^/x^)ɑ/1-ɑ
Where the “hat” (^) is used to denote a variable relative to its U.S. value, and x= n + g + d
Notice, however, that unless countries are all growing at the same rate, even relative incomes
will not be constant. That is, if the United Kingdom and the United States are growing at
different rates, then yUK/yUS will not be constant.
We make the assumption that g, i.e., the rate of technological process is identical in all
countries. This simplifying assumption is important because if g varies across countries, then
the “income gap” between countries becomes infinite.

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Figure 8(a)
First, we empirically examine how well the Solow model explains the fundamental question
of why countries are rich or poor. In Figure 8(a), we compare the actual levels of GDP per
worker in 2008 to the levels predicted by equation (8.6). Let physical capital share of ɑ=1/3
and assume that ψ=0.10. Such a value shows that each year of schooling increases a worker’s
wage by 10%. Also assume that g+d=0.75 for all countries. Lastly, we assume that
technology level, A, is the same across all countries.
The main failure of the model can be seen by the dispersion from the 45 degree line in the
above figure because technological differences are not accounted for and hence this
neoclassical model still predicts the distribution of per capita income across countries fairly
well.
Hence incorporating the actual technological levels into the analysis, we can use the
production function itself to solve for the level of ‘A’ consistent with each country’s output
and capital.
With data on GDP per worker, capital per worker, and educational attainment for each
country, we get the following figure:

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Figure: 8(b)

From the above figure, it can be concluded that the technology levels calculated using the
production function possess strong correlation with the levels of output per worker across
countries. That is, rich countries have high technology levels in addition to higher
productivity levels of human and physical capital whereas poor countries have relatively
lower technology and productivity levels.
Even though the correlation between A and levels of income is strong theoretically on paper
but it is far from perfect. “Countries such as Singapore, Trinidad and Tobago, and the United
Kingdom have much higher levels of A than would be expected from their GDP per worker,
and perhaps have levels that are too high to be plausible. It is difficult to see in the figure, but
several countries have levels of A higher than that in the United States; these include Austria,
Iceland, the Netherlands, Norway, and Singapore.
Finally, the differences in total factor productivity across countries are large. The poorest
countries of the world have levels of A that are only 10 to 15 per cent of those in the richest
countries. With this observation, we can return to equation (3.9) to make one last remark. The
richest countries of the world have an output per worker that is roughly forty times that of the
poorest countries of the world.” (Jones, 2013, p.62)

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

Similarly, in terms of education levels, rich countries outstrip the poor countries by a
relatively high margin. Workers in rich countries have about ten or eleven years of education
on average, whereas in poor countries the average education level is merely three years.
“In summary, the Solow framework is extremely successful in helping us to understand the
wide variation in the wealth of nations. Countries that invest a large fraction of their
resources in physical capital and in the accumulation of skills are rich. Countries that use
these inputs productively are rich. The countries that fail in one or more of these dimensions
suffer a corresponding reduction in income. Of course, one thing the Solow model does not
help us understand is why some countries invest more than others, and why some countries
attain higher levels of technology or productivity.” (Jones, 2013)

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)

 lOMoARcPSD|1200466

References
 Robert M. Solow (1956) - A Contribution to the Theory of Economic Growth, The
Quarterly Journal of Economics, Vol. 70, No. 1. (Feb., 1956), pp. 65-94.
 Michael P. Todaro, Stephen C. Smith (2014) – Economic Development, Twelfth
Edition
 The Solow Growth Model; Retrieved from http://www.unc.edu/~jbhill/Solow-
Growth-Model.pdf
 Debraj Ray (1998), Development Economics
 Charles I. Jones, Dietrich Vollrath (2013) – Introduction to Economic Growth, Third
Edition
 N. Gregory Mankiw, David Romer, David N. Weil (1992) – A Contribution to the
Empirics of Economic Growth, The Quarterly Journal of Economics, Vol. 107, No. 2.
(May, 1992), pp. 407-437.

Downloaded by Satish Chaudhary (grunzethedeadsoul@hotmail.com)