You are on page 1of 16


PAUL M. ROMER (1994)

Presented to: Dr. Zahid Pervaiz

Presented by: Qurra Tul Ain


Endogenous growth is different from neoclassical growth theories as the

former believes that economic growth is because of an endogenous

economic system and not because of the external forces.
Endogenous growth also reveals that the public and private sector
choices vary the rate of growth of the residual across different countries.
Two versions of the origins of endogenous growth are:
Convergence controversy
The passing of perfect competition


In recent times, there is a considerable debate on whether per capita income in

different countries is converging or not?

In this regard, Baumol(1986) used the data formulated by Maddison (1982) on
per capita income and observed the narrow gap in it between poor and rich
countries for the years 1870 to 1979.
Two objections on this observation are as follows:
Convergence only after world war II and divergence between 1870 and 1950

(Abramoviz, 1986).
Includes only successfully industrialized economies. This biasness ensures the
convergence (De Long, 1988).
As a result, data by Heston-Summers(1991) was utilized to observe
convergence in broad sample of countries. This sample showed no faster growth
in poor countries than rich ones.


Growth models explaining the convergence ignore the two assumptions of neo
classical model(Romer, 1986 & Lucas,1988)
Technological change is exogenous

Same technological opportunities are available in all over the world

So, the author considered a very simple version of the neoclassical model

Y = A(t)

Let a behavior of the economy can be summarized by the following equation

y = (1-)kk + A
y = (1-) [s- n] + A
First equation shows that the growth in output per worker is equal to

between the rate of growth of the capital-labor ratio times the

share of capital income in total income and rate of growth of output per

Second equation shows how variation in the investment rate and in the level

of output per worker should translate into variation in the rate of growth


The author used two countries, Philippines and US, to determine the

convergence of per capita in these countries.

The analysis showed that Philippines were only 10 percent as
productive as workers in the United States under the assumptions of
same technology in both countries.
It also shows that the marginal product of capital is high in Philippines
than US, so a higher rate of investment is needed in the United States
to get the same effect on output.
These results are misleading because the value of technology is not
same in both countries in 1950s.


Due to inability of Cobb Douglas production to explain the difference in

productivity among countries, the author proposed a model in which

technology was determined locally by knowledge spillovers (Romer, 1987)
following the two underlying assumptions:
Each unit of capital investment not only increases the stock of physical
capital but also increases the level of the technology for all firms in the
economy through knowledge spillovers
An increase in the total supply of labor causes negative spillover effects
because it reduces the incentives for firms to discover and implement
labor-saving innovations

The functional form of model is given as:

Yj = A(K,L)Kj1-Lj
where variables with j are ones that firm can control and those without subscript
represent economy wide
= private effect of an increase in employment on output

A(K,L) = K L-
Where = = aggregate effect of an increase in employment
Using above equation, the author found that the convergence to the mean would
take place if all the variables are held constant.

Barro and Martin (1992) explored the value of as 0.2, required to reconcile the

convergence dynamics of different states which is at slower rate. They also

note that the speed of convergence is primarily determined by the rate of
diffusion of knowledge (knowledge gap) in the presence of free movement of

Romer and Weil (1992) modify the two factor neoclassical model by
adding human capital in which the exponent of all the factors () is equal to 1/3
for their cross country regressions. Their model implies that the three-fold
increase in investment would offset 10 fold increase in output per worker in a
comparison across nations. So human capital would move from developed to
developing regions.


Everyone agrees that the conventional neo classical model with an

exponent of 1/3 on K and 2/3 on L cannot fit cross country or cross

state data.
Marginal product of investment cannot be orders of magnitudes
smaller in rich countries than in poor countries.
If the technology is same in all countries, human capital would not
move from human capital scarce regions to abundant ones and
worker would not earn higher wage if he moved to rich region.
So, this version captures only small part of endogenous growth.


This version is concerned with the slow progress in constructing formal economic

models at the aggregate level.

The evidence about growth that economists have long taken for granted and that
poses a challenge for growth theorists can be distilled to five basic facts which are
as follows:
1. There are many firms in a market economy
2. Discoveries differ from other inputs in that many people can use them at the
same time (no rivalry of knowledge goods)
3. It is possible to replicate physical activities (PF with homogeneity of degree one
in its inputs)
4. Technological advance comes from things people do (not exogenous)
5. Many individuals and firms have market power and earn monopoly rents on
discoveries (absence of pure public goods)


The neoclassical model by Robert Solow (1956, 1967) captured facts 1, 2, and 3,

but postponed consideration of facts 4 and 5.

Karl Shell (1966)proposed a endogenous growth model in which A is financed

from tax revenue collected by the government (fact 4).

Recent endogenous growth models (Neo-Schumpeterian models) emphasized

that the private sector activities contribute to technological advance rather than
public sector funding for research (facts 4 & 5).
Richard Nelson and Sidney Winter (1982) developed an alternative evolutionary

model of growth. Their appreciative theory, was flexible enough to accommodate

facts 1-5.

Nordhaus (1969) outlined a growth model which explains the

patents and monopoly power but this model is not endogenous one
as it takes exogenous technology as assumption.
Lucus (1988) included the human capital which has spillover effects.
This model accommodated facts 1-4 but left fact 5.
Whereas Romer (1986) included the expenditure on research and
development as the factor of production. This fact has spillover
effect on the level of technology. But this model assumed the
production function of degree one in all its inputs which violated the
fact 2 (research is not rival good) and fact 3 (only rival gods need to
be replicated to double output).

Two steps are required for the emergence of these models:
1. Relaxing the assumption of aggregate models with many firms having market power

(for example, a model by Avinash Dixit and Joseph Stiglitz, 1977)

2. Describing the equation with the evolution of A(t). This takes the form

A= Whereas models showing steady state growth, fill in the blank with a constant and set the

exponent equal to 1.
. The author construct models of endogenous growth in which the level of output and its

rate of growth stayed finite for all time for a range of values of that were strictly
bigger than 1(Romer, 1983; 1986).
. The difference between the models with equal to one and those with greater than 1

is the difference between studying the phase plane of a nonlinear differential equation

The economics have been progressing, starting with models based on

perfect competition, moving to price-taking with external increasing

returns, and finishing with explicit models of imperfect competition.
This paper has contributed to the convergence controversy and to an
emphasis on the exponents on capital and labor in aggregate
Neoclassical models in which market incentives and government
policies have no effect on discovery, diffusion, and technological
advance, can easily explain cross-country regressions deployed in the
convergence controversy.
By using all the available evidences, economists can move beyond
these models and try to make progress towards a complete
understanding of the determinants of long-run economic growth.

This article offers policy-makers more insight than the standard neoclassical

prescription, i.e., more saving and more schooling.

It also helps to understand the ongoing policy debates about tax subsidies;

antitrust exemptions for research joint ventures; the effects of government

procurement; the feedback between trade policy and innovation; protection for
intellectual property rights; the mechanisms for selecting the research areas
receiving public support; the costs and benefits of an explicit government-led
technology policy.
It also assists to address the most important policy questions about growth:
what are the best institutional arrangements in a developing country for gaining

access to the knowledge that already exists in the rest of the world?