Riassunto esame "development economics", Prof. Patriarca, libro consigliato "Economics of development"

Y T K L Y

the annual rate of growth of output in a time period; rt is the annual rate of growth of total

productivity also called technical progress, rk annual growth of capital, alfa and beta partial

elasticities of output with respect to capital and labor. So it is possible now to separate the

contribution of each factor.

For example let’s analyze the growth of output per head→ substract r assuming a constant

L

return to scale (β= α-1)→ r -r =r + α (r -r )

Y L T K L

R that portion of the growth of output not attributable to increases in the factors of

→

T

production (residual). It is the measurement errors, also because of resource shifts from less

productive to more productive activities.

Limitations to Cobb-Douglas function→

1. Problem in identification of shifts in the function (technical progress) from movements

along the function (changes in factor intensity)→ confusion between the two.

2. Assumption that technical progress is independent of increases in factor inputs has been

questioned.

3. This function possesses the restrictive property of constant unitary elasticity of

substitution between factors→ the function cannot represent a change in the ease of

substitution between capital and labour.

4. Criticism to the measurement of such a number of heterogeneous outputs and inputs.

Application of the Cobb-Douglas function→ first let’s consider its application in developed

countries. The function was originally conceived to see whether elasticities of output with

respect to labour and capital corresponded to factor share. Douglas observed the US

manufacturing industry curve (1899-1922), which lay between capital and labour, and

suggested to measure the effect of labour and capital on output over the period in question. →

confirmation of the relative importance of these factors of production.

then Abramovitz and Solow showed that 80-90 per cent of the growth of output per head

→But

in the US economy could not be accounted for by increases in capital per head.

So→ notion of the embodied or endogenous technical change→ capital stock series has been

adjusted to reflect changes in its composition and to allow for the fact that new additions to the

capital stock are likely to be more productive than the existing capital stock.

Cobb-douglas function→ assumption of exogenous technical change: all capital units benefit

equally from technological progress.

Production function studies of developing countries→ major conclusions of the early

production function studies of developing countries:

capital accumulation is a more important source of growth than total productivity

- growth and more important than in developed countries.

Improvement in the quality of labour are important

- Resources shifts are not so important as might been expected.

-

More recent studies:

Shaaeldin: overwhelming importance of contribution of capital formation to the growth

- process.

Young: growth miracle in the 4 east Asian countries (Hong Kong, Singapore, South-Korea

- and Taiwan)→ he used the production function approach to show that most of their

growth of output can be accounted for by the rapid growth of factor inputs.

Hu and Khan: use this approach to understand the sources of fast growth in china over

- the period 1953-94 and after the open door policy in 1978. GDP grew at 5.8 per cent per

annum and then accelerated to 9.3 from 1979 to 1994→ according to HU and Khan the

reform stimulated productivity growth through: transfer of resources from agriculture

to industry, reallocation of resources from public to private sector, encouragement of

FDI and faster growth of exports.

Senhadji: again capital accumulation is by far the most important contributor to

- measure growth. (East Asia: small contribution of TFP- total factor productivity; Africa:

negative contribution pf TFP).

Conclusions of all these studies:

Major source of growth in developing countries: increased factor inputs+ improvement

- in the quality of labour, health and education

The growth of TFP in developing countries is relatively slow→ reflection of different

- development stages.

Resource transfer form agr to ind is an important source growth, but not as expected.

-

New (endogenous) growth theory and macrodeterminants of growth→ major inspiration

behind this new theory: pioneering studies that could find no convergence of per capita

incomes in world economy (contrary to the predictions of new classical theory).

Central assumption: with not diminishing returns (so with increasing or constant) to capital,

investment is important for the long-run growth and growth is endogenous in this sense.

Lucas and Romer→ there are assumed to be positive externalities in human capital formation

(education, training) and R&D that prevent the marginal product of capital (the additional

output resulting from the use of an additional unit of physical capital) from falling and the

capital-output ratio from rising. AK model: Y= AK where alfa is 1.

→ α

First test of this new growth theory is to see whether or not poor countries do grow faster than

rich ones. If yes→ support to neoclassical theory, if not support to the new growth theory:

→

marginal product of capital does not decline. Equation: g = a+b (PCY) .

i 1 i

Gi: average growth of output per head of country I over years

PCY: its initial level of per capita income

If estimate of b1 is negative: unconditional convergence: poor countries growing faster than

rich without allowing for any difference between countries (no study has found evidence of

unconditional convergence). of b1 is invariably positive indicating divergence.

→Estimate

So new growth theory is supported by finding that education, R&D expenditure, etc,

matter, preventing marginal product from falling, so producing divergence.

Heart of new growth theory→ non-diminishing returns to capital or constant capital-output

ratio.

ENDOGENOUS GROWTH ESTIMATIONS

The level of GDP and the growth of GDP. If the Solow model is true, the higher the initial level

of GDP, the lower the rate of growth. The worst you are the faster you grow. There is a negative

relationship.

I should find b to be negative and to be significant and reliable, if b is so I can say that my

estimation accepts the hypothesis of absolute convergence.

ABSOLUTE CONVERGENCE seems not to be accepted from an empirical point of view. Cease the

coefficient b is usually not significant.

Solow model: if countries have different saving rates, they will converge in the long run, but

they will have different convergences. So I should look for conditional convergence, not

absolute one.

I can accept Solow model only if I control the saving model, I also consider that economies

different saving rates and they converge to different E. If I have higher saving model, I will

converge with a higher savings per capital. The economies converge only if they have the same

savings rate, otherwise they will converge with different E. If the economies have the same I

rates, the lower the starting point the higher the growth. Exogenous model —> Many other

variables could be determinate for the growth of countries: rate of growth of population, the

standard endogenous (education, human K accumulation), Investments in research and

development trade opens and institutions (political stability)

If the estimation of the full model gives a result b is negative, the underlined assumption of the

SOLOW model is true. Technological progress drives development and not accumulation of K.

Kaldor→ progenitor of endogenous growth theory: exogenous rate of technical progress,

which function relates the rate of growth of output per worker to the rate of growth of capital

per worker. a lower capital-labour ratio does not necessarily imply a lower capital-output

→

ratio. Countries with specialized industries do not require a higher ratio of capital output→ rich

and poor countries are not on the same production function.

Empirical studies→ in these studies only 4 variables seem to be robust: they remain

statistically significant regardless of what other variables are included in the equation.

In these studies the robust variables are:

Ratio of saving and investment to GDP

- Population growth

- Initial level of per capita income

- Investment in human capital measured by the secondary school enrolment rate.

-

Six studies are surveyed, in which the dependent variable is the growth of per capita output.

1) Barro just want to test conditional convergence, he considers the initial level of

- GDP, investment intensity and human K. He found that there is indeed convergence, b1

is negative and significant. And also Human K has an important effect is positive and

significant.

2) Mankiw he also considers the intensity of Investment.

- 3) Knight used a panel data, he adds a, b1, b2, b6, only infrastructural investment,

- trade doesn’t result to be significant. He found the there is conditional convergence.

4) Levine and Renelt —> the population growth is not reliable, all the countries that

- he puts in this equation is not significant, all the stuff about endogenous growth are not

relevant. Endogenous growth is not supported by evidences.

Trade seems to be significant, if we consider trade and monetary variable, trade is no more

robust, also trade is not affecting too much the rate of growth.

To sum up:

there is not absolute convergence, Rodrik says that if we restrict only the rate of growth

- of the manufacturing sector there is absolute convergence. What is important in the rate

of growth is the sectorial, what we produce is important.

Weakness of new growth theory:

relevance of preconditions (Institutional structure, tax system, agriculture, etc), which

- brought countries to grow at different rates.

Many models considered by this theory are closed economies without demand

- constraints.

Difficulty in characterize the growth of many developing countries by a single time trend

- because growth is volatile.

Growth diagnostics and binding constraints on growth→ a program of reforms could not

be the solution to poor growth and development performance. It is better to undertake growth

diagnostics. This framework encompasses all major strategies of development and clarifies

which strategies are effective.

Investment is the key for long-run growth, so→ understand why investment is low. Once the

diagnosis is done, effective polices follow and they must be targeted as close to binding

constraints as possible.

GROWTH ACCOUNTING

Strategy to test these theories, econometric strategy (devoted to analysis of data) that can be

useful to understand the sources of growth.

Cobb Douglas Function

The idea is that the growth rate of economy will be included by the two growth rates of capital

and labor.

Rate of growth of output = to the rate of growth of technology (technol