1.Introduction
Human
capital is an important determinant of technological progress and economic
growth of a country. The contribution of human capital to economy-wide
technological improvement through the twin channels of imitation and
innovation, and consequently, its implications on economic growth has been a
subject of much empirical and theoretical research. This paper reviews the
existing theoretical and empirical literature that links endogenous human
capital, economic growth and convergence of the growth process. The paper has
been broadly divided into four sections. Section 2 reviews the earliest models
explaining growth and convergence. Section 3 discusses the current linking
human capital and growth. Section 4 reviews the empirical literature on the
subject and Section 5 concludes.
Dear
readers
This
research paper was edited by me in International Peer Review Referred Indexed
Interdisciplinary Multilingual Monthly Research Journal –.
The
research Paper is published in the March issue of RESEARCH ANALYSIS AND EVALUATION.Journal.
To
read the published research paper, click on the weblink given below -
http://www.ugcjournal.com/assets/authors/Human_Capital_Growth_and_Convergence.pdf
2.
Growth and Convergence - The Earliest Approaches
The earliest growth models explaining
growth and convergence can be traced back to Solow-Swan (1956). Solow-Swan
model is largely characterised by a production function that exhibits constant
returns to scale and diminishing returns to each input (capital and labor) and
a constant savings ratio. In the absence of technical change, the model
predicts that due to diminishing returns to individual factors, there can be no
long-run economic growth and the economy will reach a steady state in which
there is zero growth of per capita income. Therefore, the main obstacle to
sustained economic growth is diminishing returns. If exogenous technical
progress is introduced in the basic model, then sustained economic growth is
achieved, but this is linked to the (exogenous) rate of technical change. The
technical progress overcomes diminishing returns as labor becomes increasingly
productive and, therefore, economies exhibit positive rates of per capita
income growth, which is linked to the rate of technical progress.
One shortcoming of the Solow-Swan
model is that the savings rate is exogenous and constant. Cass-Koopmans-Ramsey
(CKR) model (after the work of Ramsey (1928) that was refined by Cass (1965)
and Koopmans (1965)) is an analytical framework in which saving rate is
determined endogenously by optimizing infinitely-lived households and firms
that interact in competitive markets.
Households choose their lifetime
consumption (and savings) by maximizing their utility subject to a lifetime
budget constraint. Firms, on the other hand, choose the levels of capital and
labor they use in producing output in order to maximize profits. However, this
specification of consumer-maximizing behavior does not lead to qualitatively
different equilibrium conditions from the Solow-Swan model. The only difference
between the two setups is that, in a CKR environment, the optimal level of per
capita output in the long-run equilibrium may turn out to be lower than that in
a Solow-Swan environment.
This happens because future consumption
does not yield the same utility as present consumption due to the presence of a
discount factor. This calls for less “sacrifice” in terms of foregone
consumption and, consequently, fewer saving and a lower level of equilibrium
level of per capita income and capital-labor ratio than in the Solow-Swan model
with a constant savings rate. Akin to the Solow-Swan framework, their model too
predicts that as long as there are diminishing returns, there will be no growth
in per capita income, unless exogenous technological progress is introduced.
A key feature of the CKR model is that
the representative household plans with an infinite horizon. This assumption of
an infinitely-lived representative household is not realistic. In sharp
contrast, the overlapping generations (OLG) models introduced and studied by
Samuelson (1958) and Diamond (1965) captures the effects of finite horizons.
The OLG model assumes that individuals
live for a fixed number of discrete periods such as childhood and adulthood.
The period of adulthood for one generation overlaps with the period of
childhood for the next generation. Individuals, by assumption, do not care
about the welfare of the next generation. With logarithmic utility and
Cobb-Douglas technology, predictions of the baseline OLG model are similar to
Solow model. In the long-run, the per-capita income grows at the rate of
exogenous technical progress. For more general specifications, however, the
model may exhibits multiple equilibrium, which may be stable or unstable.
It becomes apparent from the growth
models discussed above that technical progress can overcome diminishing returns
to factors, which otherwise pose an impediment to sustained economic growth.
But technical progress in these models of neo-classical production function is
generally assumed to be exogenous, which is a restrictive assumption as it does
not give any economic explanation about how technical progress is happening.
Another alternative way to get around diminishing returns is to directly assume
a production function which is not subject to diminishing returns. AK model
does the same by assuming that output is a linear function of capital, where A
is the index of technology and K is defined in a broad sense to include
different forms of capital stock such as human capital, physical capital,
environmental capital etc.
An economy characterized by such a production
function will accumulate capital continuously without experiencing diminishing
returns to it and therefore, can experience sustained economic growth
prospects. An alternative approach is to introduce additional inputs to the two
basic inputs (capital and labor), which has been developed by Mankiw, Romer and
Weil (1992). In their model, an additional form of capital, human capital, or
the stock of knowledge is used in production. The introduction of this form of
capital does not change the main findings of the Solow model. Human capital,
however, can generate sustained economic growth. This subject has been
discussed in detail later in the next section.
Another approach to overcome
diminishing returns to factor is postulated by the learning by-doing and
capital spillovers model introduced by Arrow (1962) and developed by Romer
(1986). Learning-by-doing model functions on the assumption that the process of
investing in physical capital by firms simultaneously increases their
efficiency in production. This positive effect of experience on productivity is
called learning by doing. Knowledge is considered to be a public good and
therefore, investment in physical capital by a single firm has spillover
benefits that raises productivity of all the firms in the economy.
A single firm’s investment in physical capital
increases aggregate physical capital stock and generates positive spillovers
that eliminate the tendency for diminishing returns to capital. Thus, the
learning-by-doing and spillover effects yield sustained economic growth in the
economy. However, an implication of these models is that decentralized outcome
is nonoptimal as individual firms do not internalize the positive spillovers
generated by the process of physical capital accumulation and therefore, do not
invest enough in physical capital stock.
Thus, there is scope for intervention
by the government in the form of incentives to spur physical capital
accumulation, that raises returns to physical capital and ensures that social
optimum is reached in the decentralized economy. R. J. Barro (1990) postulates
a model of public infrastructure in which provision of public services by the
government eliminates the tendency for diminishing returns to capital
accumulation. The purchases of goods and services by the government enters into
the production function of a firm as a pure public good. The production
function exhibits diminishing returns to private physical capital but constant
returns to scale with respect to private capital stock and the flow of public
services provided by the government. This constant returns to private physical
capital and public services together generate sustained economic growth.
A second way to achieve sustained
economic growth in the long-run is by improvement of production process through
technological improvements. Technological improvements can happen through the
development of new intermediate inputs that neither complement or substitute
existing intermediate inputs as postulated by Romer (1990) in his product
variety model. Alternatively, technical improvements can occur through quality
improvement of existing inputs that render the older inputs obsolete as postulated
by Aghion and Howitt (1992) in their quality ladders model. Both models
generate sustained economic growth in the long-run.
One of the main questions that all
these growth theories attempt to answer is whether poor countries are likely to
catch up with rich countries. In other words, will the per capita income of
poor economies converge to the per-capita income of rich economies? These
initial growth theories explain convergence as a result of diminishing returns
to inputs which are being used in production. As long as physical capital
exhibits diminishing returns, the poor economies that have lower initial
capital per worker have higher rates of return and therefore, higher growth
rates as compared to rich economies that have higher initial capital per
worker. As a result, poor economies will converge with rich economies to the
same steady state without conditioning on any other characteristics (such as
savings ratio, population growth) of economies. This process of convergence is
referred to as absolute convergence. This is true when the economies are
structurally similar. In comparison with this, the process of convergence is
conditional when an economy with a lower initial per-capita capital stock grows
at a faster rate and converges to its own steady-state depending upon the other
characteristics of the economy. Since real economies tend to be structurally
different, it is conditional convergence that has found much greater empirical
support.
The most well known initial empirical
study of absolute convergence is by Baumol (1986). He finds no evidence of
absolute convergence for a sample of 72 countries. Kormendi and Meguire (1985)
and Grier and Tullock (1989) test for conditional convergence and their
regression results provide evidence of conditional convergence. These initial
convergence studies do not include human capital as an explanatory variable,
which is a major drawback. R. J. Barro, Sala-i Martin, Blanchard, and Hall
(1991) have proposed the concept of beta (â) convergence which is a popular
methodology of investigating convergence empirically. â-convergence considers
whether the growth rates of countries exhibit a negative correlation with the
initial level of real GDP per worker.
If they are negatively correlated,
this implies that countries with low real GDP per worker possess more potential
for faster growth rates than countries with high real GDP per worker. To study
convergence, R. J. Barro et al. (1991) includes the initial income variable in
his regressions. He reports absence of absolute convergence in a broad sample
of 98 countries. He finds that the coefficient of initial income turns
negatively significant when the initial measures of human capital are included.
This leads R. J. Barro et al. (1991)
to conclude that the data support conditional convergence. He also finds that
the measures of human capital are positively and significantly related to
income growth rates. Human capital has been extensively modeled in endogenous
growth theories as an input that helps in countering the diminishing returns of
neo-classical production function.The next section discusses in some detail the
role of human capital in explaining growth and convergence.
3.Current
Approach linking Human Capital and Growth
Two different approaches have been followed
in the endogenous growth literature to model the relationship between human
capital and economic growth. The first approach has been postulated by scholars
like Lucas Jr (1988) and Rebelo (1991). Here, human capital is a direct factor
of production, which is positively related to output growth just like other
factors, such as physical capital and labor. Human capital accumulation implies
acquisition and up-gradation of skills by the work force that increases the
productivity of workforce, which ultimately culminates into higher economic
growth rates. Although,
there exist diminishing returns to each factor individually, there are constant
returns to physical and human capital together. This property of the production
function prevents the marginal product from falling as human capital and
physical capital are accumulated, and this gives the model the sustained growth
property. Here, the rate of growth depends upon the rate of accumulation of
human capital. Since they formulate an AK-type model structure in equilibrium,
an implication of their model is that economies that differ in their initial
conditions (different initial capital-labor ratio) will grow at different rates
indefinitely and will never converge.
The second approach that has its origin
in the contribution of Nelson and Phelps (1966) and Benhabib and Spiegel (1994)
focusses on the relation between human capital and technological progress.
Under this approach, human capital does not enter the production process
directly but facilitates the adoption and development of technology (sometimes
differentiated by imitation and innovation activity as two distinct routes for
technological progress). This strand of literature de-emphasises the role of
human capital accumulation and highlights the importance of technological
progress. Within this framework, the economic growth rate is determined by the
rate of innovation (and/or imitation) and therefore, subsequently, by the level
of human capital and not by the rate of human capital accumulation. Romer
(1990) formulates an endogenous growth model in which he makes explicit the
role of human capital in promoting technological progress and, therefore,
growth.
According to Romer (1990), new technologies
are developed by inventing new intermediate product varieties. There are two
distinct roles of human capital. One is for improving technology (that is, by
developing new intermediate product varieties), and the other is for final
production (that is, by increasing productivity). Aghion and Howitt (1992) have
formulated a slightly different framework in which new technology is developed
by improving the quality of existing intermediate inputs. Invention of a
higher-quality intermediate good renders previous intermediate goods as
obsolete.
These endogenous growth models predict
a pattern of convergence across economies based on the diffusion of technology
from leader to follower countries, instead of diminishing returns to capital as
predicted by the neo-classical models. The researchers in leader countries
expend effort in innovation. Innovation occurs either through the production of
new intermediate varieties or through quality improvements. The follower
country does not innovate but imitates and adapts the intermediate inputs
produced in the leader country. The cost of imitation is less than the cost of
innovation when only a small proportion of new ideas have been copied, but it
increases as the pool of uncopied ideas contracts.
This cost structure implies a kind of
diminishing returns to imitation and, thereby, generates a pattern of
convergence. The follower country will have a higher growth rate than the
leader until it has managed to emulate and adopt all the intermediate goods
that have been developed by the leader. After that point, there will be
simultaneous adoption of all intermediate goods that are developed and both the
countries will grow at the same economic growth rate ( R. Barro (1995) ch.8).
Benhabib and Spiegel (1994) discuss the link between human capital and
technology diffusion. Building on the formulation by Nelson and Phelps (1966),
they develop a model in which a higher stock of human capital spurs technical
progress in the long-run.
They postulate that the level of human
capital affects the total factor productivity through two channels. Firstly,
human capital directly influences productivity by enabling a country to
innovate new technological capabilities suited to domestic production.
Secondly, they assume that the ability of a country to adopt and adapt new foreign
technologies (that is, the catch-up effect) depends upon its domestic human
capital stock. Here the economies converge to world technology frontier, and
eventually, also in terms of economic growth rates. The follower, which is at a
distance from the world technology frontier, will have a higher growth rate
than the leader due to the catch-up effect.
Once the follower gets closer to the
world technology frontier, the catch-up effect vanishes and the leader and
follower will grow at the same rate at the world technology frontier. The role
of human capital is crucial as it is the human capital stock that determines
the strength of the catch-up effect. A follower with a higher human capital
stock adopts new technology and converges to the technology frontier at a
faster pace as compared to a follower country with a lower human capital stock.
Using cross-country data from 78
countries over the period of 1965 to 1985, Benhabib and Spiegel (1994) estimate
a positive relation between human capital stock and economic growth. Similarly,
Pritchett (2001) finds a positive relation between human capital and economic
growth. However, the findings of Benhabib and Spiegel (1994) and Pritchett
(2001) have been criticized by other researchers citing reasons of misspecification
of model and measurement errors (see, for instance, Topel (1999); Temple
(1999). In another influential study, Krueger and Lindahl (2001) observe that
human capital enhances growth only for the countries with the lowest level of
education. That is, education matters only for catching up but not for
innovation at the frontier. Also, they find evidence in support of conditional
convergence.
In an attempt to resolve this
Krueger-Lindahl puzzle, Vandenbussche, Aghion, and Meghir (2006) argue that human
capital does not affect innovation and imitation uniformly. They develop an
endogenous growth model, where innovation makes relatively more intensive use
of skilled labor and imitative activities make relatively intensive use of
unskilled labor. Thus, in their model, human capital is skill differentiated.
They show that human capital affects the rate of technical progress via a level
effect and a composition efffect. Holding the composition of human capital
constant, an increase in the stock of human capital is always growth-enhancing.
However, holding its level constant,
the growth-enhancing properties of human capital depend on both its composition
and the distance to the technological frontier. The growth-enhancing impact of
skilled labor increases with a country’s proximity to the world technology
frontier, where proximity is measured by the ratio between the total factor
productivity in the country and the corresponding variable for a frontier
economy such as the US. Conversely, the growth-enhancing impact of unskilled
labor decreases with the proximity to the world technology frontier. Their theoretical results
state that tertiary education should become increasingly important and primary
and secondary education less important for growth as a country moves closer to
the technology frontier. Using a panel dataset covering 19 OECD countries for
period 1960-2000, they find evidence in support of their theoretical findings.
Ang, Madsen, and Islam (2011) empirically investigate the predictions of the theoretical
model of Vandenbussche et al. (2006) for developing countries. In particular,
they investigate whether the contribution of human capital to productivity
growth depends on the composition of human capital and proximity to the
technology frontier in a panel of 87 sample countries over the period from 1970
to 2004.
Their results show that the growth
enhancing effects of tertiary education attainment or skilled human capital
increase when high and medium income countries move closer to the technology
frontier. Human capital is not contributing to growth in low income countries,
suggesting that they neither innovate nor imitate. Also, they find evidence of
technology convergence independent of human capital in low income countries,
implying that being far from the frontier allows one to experience faster TFP
growth. Income convergence is the joint outcome of the twin processes of
capital deepening and technological catch-up. Since TFP is the closest measure
of technology, researchers have investigated whether countries have come closer
in terms of TFP levels.
This has given rise to the concept of
TFP-convergence. Income convergence across countries gets either accelerated or
thwarted depending on whether initial TFP-differences narrow or widen over time
(N. Islam, 2003). The main drawback of the study by Vandenbussche et al. (2006)
is that they assume that there exists an exogenously given composition of
skilled-unskilled human capital. They consider only the benefit of skilled
labor and ignore the fact that skill acquisition is not cost less.
Besides this strand of literature
linking human capital, technical progress and economic growth, there exists
another line of literature that describes the process of human capital
formation as a source of demographic transition from a Mathusian economy to a
Modern economy. The genesis of this
strand of literature can be traced back to the seminal work of Becker (1960)
where he mentioned the concept of “child quantity-quality trade-off” for the
first time. Becker defines child quality as the expenditure incurred on a
child. So, higher the expenditure incurred on a child, higher is the skill-set
of a child. Parents maximize their utility which depends on child bearing and
own consumption subject to a budget constraint to determine the optimal
quantity and quality level of children. In this particular setting, Becker
proposes that parents experience a trade-off between the quantity and the
quality of their children as per capita income rises. Parents start spending more
on children’s education and bear a lower number of children, which leads to a
decline in fertility.
The literature linking human capital
formation and demographic transition of an economy highlights the process of
human capital accumulation as a trigger for the child quantity-quality
trade-off, which leads to the transition of an economy from a primitive economy
having high fertility and low economic growth to a modern economy with low
fertility and higher economic growth. The central idea behind theories
belonging to this line of literature is that technical progress leads to better
utilization of resources, which in turn, leads to higher wages.
However, higher technical progress
requires skilled labor. Therefore, altruistic parents who care about their
children focus on increasing the human capital investment in their children.
This triggers a child quality-quantity trade-off wherein parents prefer having
fewer but higher quality/more educated children. As a result, economic growth
is accompanied by a fertility transition from high to low fertility. Becker,
Murphy, and Tamura (1990) formulate a theoretical model with endogenous
fertility to characterise Malthusian economies and modern economies. They
assume that rate of return on human capital rises as the stock of human capital
increases. The reason behind this assumption is that human capital is largely
defined by knowledge embodied in individuals. The benefit of imparting
additional knowledge to an individual depends positively on the knowledge he/she
has already gained. Becker et al. (1990) explain this rationale by citing an
example that learning of complicated mathematical concepts is easier when there
is conceptual clarity of basic mathematical concepts. Therefore, when human
capital stock is abundant, families have lesser children and invest more in
their offspring as rate of return on human capital is high. On the other hand,
individuals have larger families and invest little in children when human
capital stock is limited. This assumption generates two steady states in the
model. One is the Malthusian steady state with high fertility, lower stock of
human capital and lower return on human capital and therefore, lower per capita
income. The other one is the developed steady state with low fertility and
higher human capital stock which yields higher returns on human capital
investment and higher per capita income. They also show that an economy may
switch from a Malthusian trap to modern economic growth after a threshold value
of human capital accumulation.
A pioneering work in this field is the
“Unified Growth Theory” postulated by Galor and Weil (2000).”Unified Growth
Theory” is a comprehensive endogenous growth model of technology, fertility and
human capital which explains the entire evolution process of mankind starting
from a Malthusian economy through a Post-Malthusian regime, to a demographic
transition and eventually the period of sustained economic growth. It explores
the interlinkages among technical progress, per capita income and human capital
formation process. The impact of technical progress on per capita income and
the child quantity-quality trade-off determines the entire process of economic
growth. Galor and Weil (2000) postulate a technical progress function that
depends on population size and education. In the Mathusian regime, the economy
is in a Malthusian trap with low per capita income. The technological progress
occurs at a slow pace such that the rise in income per capita is always offset
by popluation growth. In the intermediate stage of Post-Malthusian regime, the
economy takes off due to higher rate of technical progress caused by the
increase in the size of population during the Malthusian regime. A higher rate
of technical progress increases the relative return to human capital. This
triggers a child quantity-quality trade-off where parents start spending on
their offsprings’ education and have lesser number of children. This induces a
demographic transition in which fertility rates decline. This eventually the
paves way to the period of sustained economic growth. Galor and Moav (2002)
extend Galor and Weil (2000) by introducing heterogeneous preferences of
individuals about quantity and quality of children.
They assume that these preferences
about child quantity-quality are hereditary implying that if an individual
prefers quality over quantity of children, his future generations will share
same preferences. Thus, the population is segregated into different groups
according to their choices between child quantity-quality. Their model
postulates that population composition is reshaped due to changing
technological and economic conditions such that people with quality-biased
preferences survive. As a result, human capital accumulates leading to faster
technical progress which leads to the transition from the Malthusian era to a
demographic transition and eventually, paves the way to sustained economic
growth.
In a similar vein, Moav (2005)
develops a theory of fertility and child educational choice to explain the
persistence of poverty across countries. He assumes that the cost of education
is in terms of the wages foregone on account of the time spent on educating the
children. Further, he assumes that individuals’ productivity in educating
children increases with their own human capital whereas the child rearing costs
are equal across all individuals. A lower level of parents’ education (lower
human capital) implies that the parents’ time is cheaper, and therefore, it is
cheaper to have children. As a result, parents have more children and incur
lesser investment in human capital of their children, which leads to lower per
capita income. The high fertility rates further dilute the accumulation of per
capita physical capital which reinforces the impact of child quality choice on
economic development.
Thus, households in poor economies
choose higher fertility rates with lower investment in their offspring’s
education and lower levels of capital transfers; and therefore, poverty
persists. In contrast, families in the richer countries choose lower fertility
rates with higher investment in education and higher levels of capital
transfers, and therefore, high income persists in such economies. Thus, the
model offers explanation for cross-country output differences and for the phenomenon
of club convergence.
Club convergence is a weaker form of
convergence in which countries having the same structural characteristics and
similar initial conditions converge to similar levels of per capita income;
that is, poor countries and rich countries converge to low and high income
levels respectively. Countries in the club of the rich converge to a high
income-per-capita steady state, whereas countries in the club of the poor
converge to a low-income level. The poor countries fail to catch up with the
rich because of insufficient progress in education, which is due to high
fertility rates.
From the theoretical and empirical
studies reviewed so far, it can be concluded that education has far-reaching
impact on the growth prospects of a country. However, as pointed out by
empirical studies like Benhabib and Spiegel (1994), Pritchett (2001), schooling
per se does not necessarily lead to higher growth. Quality of schooling also
matters for economic growth. One of the reason for divergent growth
trajectories of developing countries vis-´a-vis developed countries is the
significant qualitative differences in their education systems. The next
section discusses the empirical
literature on the linkages between human capital, economic growth and
convergence.
4.
Human Capital, Growth and Convergence-Empirical Literature
In the existing empirical literature
on human capital, growth and convergence, human capital stock is largely
measured using various measures of schooling such as mean years of schooling
attained, net enrolment ratios. These are imperfect measures of human capital
stock as they measure only the quantity of schooling, not the quality. The
quality of schooling varies substantially across countries. Presently, in
practice, there are two approaches followed for measuring the quality of human
capital. The first includes measures of schooling inputs, such as expenditures
per student, pupil-teacher ratio, and teachers’ salaries etc.
The second refers to the direct
measures of cognitive skills such as science, mathematics and reading scores on
internationally standardized tests of cognitive skills. Lee and Barro (2001)
have compiled test scores on examinations in science, mathematics and reading
tests for students of various age groups in different years for 58 countries.
Using this dataset, they investigate the determinants of quality of human
capital. Their regression results show that family factors (such as income and
quantity of schooling) and school inputs (pupil-teacher ratio, average teacher
salary and length of school year) are closely related to school outcomes, as
measured by internationally comparable test scores, repetition rates and
dropout rates. Their study concludes that school inputs and family factors play
a major role in improving school quality.
R. J. Barro (2001) analyzes the
determinants of growth in an unbalanced panel of about 100 countries for the
period of 1965-1995. He finds evidence in support of conditional convergence.
The growth rate of per capita GDP is inversely related to the initial level of
per capita GDP, keeping the influence of measures of government policies,
institutions and character of national population constant. Growth is
positively related to the initial level of average years of adult-male educational
attainment at secondary and higher levels. He also analyzes the relationship
between quality of human capital and growth for a smaller unbalanced panel of
43 countries. He uses data on three test scores (science, mathematics, and
reading) as indicators of quality of human capital along with a measure of
quantity of human capital . (male post-secondary schooling) in the growth
regression. He finds that science scores have a positive and significant effect
on growth and, in terms of magnitude, its effect is more important than
educational quantity. Mathematics scores are also a significant determinant of
growth and the magnitude of this effect has been found to be larger than that
of the science scores. Finally, reading scores turn out to be an insignificant
determinant. The overall result is that the quality of schooling is far more
important for growth than the years of schooling.
Hanushek and Kimko (2000) provide an
extensive discussion of how scores from cognitive skill tests can be used to
measure the quality of human capital and its effects on economic growth. They
use data from six voluntary international tests of mathematics and science over
the period 1964-1991 to build a measure of quality of education. Four of these
tests are organized by the International Association for the Evaluation of
Educational Assessment (IEA) and two tests were organized by the International
Assessment of Educational Progress (IAEP). Hanushek and Kimko (2000) find that
adding educational quality to a base specification, including only initial
income and educational quantity, increases the explanatory power of the model
from 33 to 73 percent. The effect of years of schooling is greatly reduced by
including quality, leaving it mostly insignificant. At the same time, adding
the other factors leaves the effects of quality basically unchanged. The
hypothesis of conditional convergence is supported by their results as the
coefficient of initial income is negative in all the specifications. Several
studies have since found very similar results. Extensions of the measure of
Hanushek and Kimko (2000) have been used in the cross-country growth
regressions by Bosworth and Collins (2003) and in the cross-country
industry-level analysis by Ciccone and Papaioannou (2009). Both of these also
find that educational quality strongly dominates any effect of educational
quantity on growth. Hanushek and Woessmann (2012a) extend previous measures of
Hanushek and Kimko (2000) to improve direct comparisons of student knowledge
over time, across tests, and across age groups. The new data comprises 77
countries, and observations are updated up to 2003. They have repeated the
cross-country growth regressions of Hanushek and Kimko (2000) for the expanded
set of countries. Their estimate of human capital quality was found to be
positively significant signifying that a one standard-deviation increase in
test-scores would increase the long-run growth rate by two percentage points.
Hanushek and Woessmann (2012a) look at
the distribution of scores by defining two variables that measure the
proportion of students that meet a threshold level of achievement. The first
was a score of 400 or above on the transformed international scale, that is,
one standard deviation below the mean test scores for OECD countries (meant to
capture basic literacy) and the other 600 or above (to capture high
achievement). The estimates for the two threshold levels were highly
significant indicating that both basic and high achievement are important
determinants of growth, with the coefficient on high achievement substantially
greater than the coefficient on basic skills. The effect of the basic-literacy
share does not vary significantly with the initial level of development, but
the effect of the high achieving share of students is significantly larger in
countries that have more scope to catch up to the most technologically advanced
countries. These results appear consistent with a mixture of the basic models
of human capital and growth mentioned earlier. The accumulation of skills as a
standard production factor, emphasized by augmented neoclassical growth models
(e.g., Mankiw, Romer, and Weil (1992)), is probably best captured by the
basic-literacy term, which has positive effects that are similar in size across
all countries. But, the larger growth effect of high-level cognitive skills in
countries farther from the technological frontier is most consistent with
technological diffusion models (e.g., Nelson and Phelps (1966)). Their results
give consistent support for the hypothesis that quality-adjusted human capital
and its interaction with the technology gap are essential for growth.
5.
Conclusion
A broad conclusion that can be drawn
from this review of literature analyzing the relation between human capital and growth is that quality of
human capital also matters for growth along with the quantity of human capital.
As the existing growth literature reveals that human capital is a major driver
of economic growth, therefore, investment in human capital has been a primary
focus of development policy worldwide. While there has been a significant
progress in expanding access to education but it has not led to a concomitant
improvement in learning outcomes among children in many countries. Studies
assessing learning outcomes among school children across low- and middle-income
countries have consistently found that effective learning in schools in these
countries is abysmally low (Pritchett, 2013; Snilstveit et al., 2015). As P.
Glewwe (2013) point out, school enrolment is not the sole objective of
education policy but instead, the actual intent is to prepare the present
generation for a better future by honing their basic and advanced skills.
Therefore, there is a need to reconsider the role of schooling in the process
of economic growth by focussing on the qualitative aspects of schooling and its
consequent impact on economic growth of a country.
Swati
Saini
Assistant Professor, University of Delhi, New Delhi India, Email:: swatisaini1128@gmail.com