Endogenous Growth Theory

and
Growth Policies
Introduction
• Solow model is about how GDP and GDP growth are determined by the savings rate, rate of
population growth, and the rate of technological progress.
• The question analyzed in the EGT is “How do society’s choices affect these parameters?”
• In many developed countries, invention and advances in technology are the key determinants of
growth.
• Technological advances are much less important for poor countries → it’s more important to invest in
human and physical capital and borrow technological advances from others.
• Endogenous growth theory (Romer, Lucas) explains how society’s choices lead to technological
progress and growth.
Trouble with neoclassical growth theory
• By the late 1980s, there was great dissatisfaction with the neoclassical growth theory since:
1. It does not explain the economic determinants of technological progress.
2. It predicts that economic growth and savings rates are uncorrelated in the steady state.

• Endogenous growth theory emphasizes different growth opportunities in physical capital and
knowledge capital.
• There would be diminishing marginal returns to physical capital, but perhaps not to knowledge capital.

• The idea that increased investment in human capital increases growth is key to linking higher savings
rates to higher equilibrium growth rates.
Nobel Prize 2018: Paul Romer (endogenous growth theory) shared
it with William Nordhaus (economics of climate change)

Many countries prosper due to technological progress-led economic growth, and many others lag behind because they
are technology scarce.

Solow’s long-run growth model: Y=AF(K, L). Solow treats A, ideas or technology, as a public good and exogenous
(determined outside the model).
Solow’s assumption: diminishing returns to capital.

Romer’s assumptions: production function is Y= aK. Technology is endogenous, and constant returns to capital.

Constant MPK means that if total factor productivity remains constant, increasing both capital and labour by a given percentage will increase output by that same percentage.
Diminishing MPK means increasing capital while keeping everything else the same will initially increase overall output but will generate less output the more that capital is
increased. As capital increases, output increases, but at a decreasing rate → diminishing MPK.
Mechanics of Endogenous Growth
• Need to modify the production function to allow for self-sustaining,
endogenous growth.

• Figure 1 shows the Solow growth diagram, with the steady state
equilibrium at point e where investment (=saving) equals
depreciation.

• If investment is above depreciation, economy is growing as more

capital is added → the process continues until sy =dk. Fig. 1. Solow model
• The economy is illustrated in Figure 2 (Romer model) and described by
a production function with a constant MPK: Y=aK, vis-a-vis diminishing
marginal product situation in Figure 1 (Solow model).

• Production function and investment curve become straight lines and

are always greater than depreciation→ the higher the investment
rate, the bigger the gap between investment and depreciation = faster
the economic (GDP) growth.

Fig. 2. Romer model (EGT)

Mechanics of Endogenous Growth
• If the savings rate, s, is constant and there is neither population growth nor depreciation of capital,
then the change in the capital stock is defined as:

K = sY = saK
OR (2)
K
= sa
K
→Growth rate of capital is proportional to the savings rate
Y
• Output is proportional to capital, thus the growth rate of output is = sa (3)
Y
→ The higher s, the higher the growth rate of output
Deeper economics of endogenous growth
• Eliminating diminishing marginal returns to capital runs against prevailing microeconomic principles
• If there are constant returns to capital alone, there will be increasing returns to scale to all factors
taken together → larger and larger firms become increasingly efficient, and should see a single firm
dominate the entire economy
• It’s not realistic, so we need to eliminate the possibility of increasing returns to scale to all factors,
and constant returns to a single factor
• Alternatively, a single firm may not capture all benefits of capital → some external to the firm
(Romer)
• When a firm increases K, firm’s production increases, but so does the productivity of other firms
• As long as private return has constant returns to all factors, there will be no tendency towards
monopolization of a firm.
Private vs. Social Returns to Capital
• Within his endogenous growth model, Romer distinguished private returns of capital from social returns
of capital.
• Investment produces not only new machines, but also new ways of doing things.
• Firms DO capture the production benefits of a new machine (PRIVATE RETURNS)
• Firms may NOT capture the benefits of new technologies and ideas, since they are easy to copy (SOCIAL
RETURNS)
• Endogenous growth theory depends on the notion that there are substantial external (SOCIAL) returns
to capital.
• Not realistic for physical capital, but quite right for human capital:
I. Contribution of new knowledge is only partially captured by the creator
II. From one new idea springs another → knowledge can grow indefinitely
 Ideas, knowledge, innovation…

Ideas are non-rivalrous (consumption of one does not prevent simultaneous consumption of others,
e.g. broadcast TV), and cannot be sold in competitive markets because marginal cost is zero (cost of
providing it to an additional individual is zero).

Ideas also create spillovers. Spinoffs and duplicates can relatively easily be made once a product
with a new idea is public or sold in the market.

Very few markets have CocaCola kind of secret recipes/ ideas.

Does capital have diminishing returns or not?

Depends on definition of “capital.”

If “capital” is narrowly defined (only plant & equipment), then, yes.

Advocates of endogenous growth theory argue that knowledge is a type of capital.

If so, then constant returns to capital is more plausible, and this model may be a good description of
economic growth.
What the Romer model says (contd.)

As discussed, Romer (1990) created a growth model based on ideas and innovation privately produced by for-
profit firms in monopolistically competitive markets with spillovers.

Growth happens when there is a spurt in new knowledge, ideas and innovation, and most R&D is produced by
profit-seeking firms.

Solow did not explain where technological progress would come from.

Romer endogenized technological progress and determined it within the model to explain long-run growth. Left
alone, markets would produce too few ideas and hence, government investment in R&D and innovation is
necessary.
Convergence
• Do economies with different initial levels of output eventually grow to equal standards of living or
converge?
• Neoclassical growth theory predicts absolute convergence for economies with equal rates of saving
and population growth and with access to the same technology → should all reach the same steady
state level of income
• Conditional convergence is predicted for economies with different rates of savings and/or population
growth → steady state level of income will differ, but the growth rates will eventually converge
• Endogenous growth theory predicts that a high savings rate, particularly, investment directed toward
human capital and innovation, leads to a high growth rate.
Convergence
• Do economies with different initial levels of output eventually grow to equal standards of living or
converge?
• Robert Barro tested these competing theories, and found:
1. Countries with higher levels of investment tend to grow faster
2. The impact of higher investment on growth is however transitory

Countries with higher investment will end in a steady state with higher per capita income, but not with a
higher growth rate.
Countries do appear to converge conditionally, and thus, endogenous growth theory is not very useful for
explaining international differences in growth rates.
Growth Policies

• Population growth and Malthus

• Lessons from the Asian Tigers
• The Chinese Growth miracle
• The truly poor countries
• Natural resources
• Social infrastructure (political institutions)

• New growth theories versus the old

Solow Model with Endogenous population Growth

• One of the oldest ideas in economics is that population growth works against the achievement
of high income (Thomas Malthus, 1826).

• The Solow growth model predicts that high population growth, n, means lower steady state
income as each worker will have less capital to work with.

• Over a wide range of incomes, population growth itself depends on income.

 • Very poor countries have high birth rates and high death rates, resulting in moderately high
population growth.

• As income rises, death rates fall (medical facility is up) and population growth increases.

• At very high incomes, birth rates fall, some even approaching zero population growth (ZPG).

Examples:
High population growth in poor countries, some African countries
Very low population growth in advanced European countries like Germany
Old-age-care-of-parents rule by Assam govt.
The growth story of the Asian Tigers

• South Korea, Singapore, Taiwan and Hong Kong have traditional Solow type growth experience,
except that their high long-run growth has not been due to higher technological progress (TFP).

• They have high saving and investment rates, increased labour and capital inputs, and raised
education (human capital).

• Growth in TFP (tech progress) is high but not remarkable.

• Things these countries have in common: relatively stable governments, open progressive pro-
free-market economies, thrust on raising exports, higher competitiveness.
The Chinese miracle
• High saving rate, high capital per worker, high output per worker
• High productivity,
• Moderate or low population growth
• Export-oriented economy, also technological rip-offs
Poor countries’ plight and low long-run growth

• Low per capita income, low saving rate

• High population growth
• Extremely low rate of technological progress
• Extremely low rate of technological diffusion
• Lower exports, higher debt
• Lower labour productivity (output per worker)
• Lower human capital per worker
• Lower physical capital per worker
• Low-quality political institutions
Truly Poor Countries
• Ghana, and many other countries, experienced very little growth in recent years
• Income is so low that most of the population lives on the border of subsistence
• Can the Solow growth model explain these countries’ experiences? YES
• Savings in Ghana is quite low (9.3% of GDP vs. 34.3% and 19.4% of GDP in Japan and the US
respectively)
• Population growth is very high in Ghana and other poor countries relative to the US and Japan
• The effects of low savings rates and high population growth rates are as predicted by the Solow growth
model: low levels of income and capital per capita.
Natural resources

• One day, natural resources (finite) will be depleted.

• Technological progress lets us produce using fewer resources (e.g. energy efficiency of room
lighting has increased by a factor of 4500 since Neolithic times – William Nordhaus 1997)

• As specific resources come in short supply, prices will rise, leading producers to shift to substitutes
(e.g. petroleum -> biofuel, solar energy and renewable energy etc.)

• Natural resources are linked with environmental protection.

• Non-market activity, politics and public policy. (e.g. legislation on natural resources extraction and
surrounding domestic and global politics)
Social infrastructure (political institutions)

• Sound legal system (proper enforcement of rule of law)

• Stable and progressive tax system
• Limits on bureaucratic power
• Political competition and stability
• Ease of doing business (pro-free-market regulation)
• Civic engagement and public trust in society
• Culture and norms (e.g. UN Diplomatic parking tickets and traffic violations in NYC
(Fisman and Miguel 2009))
The role of political institutions and history in
economic growth and development
Minimum reading required

• Acemoglu, Johnson and Robinson 2001

• First 8 pages (upto the starting of section II, data) and conclusion section

• Banerjee & Iyer 2005

• First 9 pages (upto the starting of section III. Data, p. 1199 has 2-3 lines remaining) and conclusion section

• Iyer 2010
• First 5 pages (upto the starting of section IV, data) and conclusion section
Role of institutions and history
• Colonial settlers → Past Institutions
• Past Institutions → Current Institutions → incentives
→investment → growth
https://ourworldindata.org/economic-growth
What does cause large per capita income differences across
countries?
What does constitute ‘Institutions’?

Why do they matter? E.g. North and South Korea, Nigeria and Benin, or more closely, India and
Pakistan
• cross-country institutions and development (AJR, 2001, 2004; Hall and Jones, 1999;
Knack and Keefer, 1995)
• property rights and investment/output: Besley (1995)

Nature of institutions, property rights, distortionary policies cause differences in investment in

physical and human capital and hence differences in factor efficiencies and income (Engerman and
Sokoloff, 2000)
 Institutions: definitions

• Douglas North (1990): Institutions are the rules of the game in a society or, more formally, are the
humanly devised constraints, formal and informal, that structure human interactions. Together they
define the incentive structure of societies and specificially economies. 1) rules, laws, constitutions, 2)
norms, behavior, conventions… or simply the rules of the game.

• Avner Greif (2006): An institution is an integrated system of beliefs, norms, organizations and rules that
generate a regularity of behavior. Man-made, non-technological features that influence economic
outcomes by constraining behavior.

• Examples: Formal constraints: laws of the land, constitution

• Informal constraints: caste system (society controls norms informally)
Institutions: definitions

• Daron Acemoglu and James Robinson (2012, Why Nations Fail): “Economic institutions shape economic
incentives, the incentives to become educated, to save and invest, to innovate and adopt new technologies,
and so on. Political institutions determine the ability of citizens to control politicians.”

• Essentially, institutions are a catch-all phrase for the systems we create as a society which ultimately
determine the structure of market interactions.

• Institutions affect efficiency and productivity by influencing and coordinating:

consumption,
savings,
exchange,
investment,
production and
innovation.
Institutions can be differentiated according to who implements them

• Conventions (individuals enforce it themselves) (e.g. offering own seats to the elderly [/ mother with a
baby in arms] in the bus)

• Ethical rule (individuals self-commit to ethics) (e.g. samaritan help for road accident victims)

• Customs, norms (society controls informally) Customs/ norms (e.g. expecting people to enter any
place of worship without shoes on, following traffic rules in places where traffic police is not around)

• Private rule (organized private enforcement) (e.g. private job/ industrial contracts)

• State law (organized state enforcement by government) (e.g. law of the land, constitution, government
rules)
 Examples of institutions
• Political systems: Democracy, dictatorship, communism, media, corruption.
• Market systems: Socialism, capitalism/corporatism, free markets.
• Property rights: Goods, land, radio frequencies.
• Judicial Systems: Adversarial judicial system versus investigative judicial system, bankruptcy
protections.
• Culture: Social norms, public trust, freedom, individual rights, religion, caste system.
• Conflict: War, theft, insurgency.
Examples:

1. brief political history and evolution of institutions in India/ USA, North Korea, China, media bias and media freedom, corruption

2. India is increasingly perceived as a free market mixed-economy, moving away from socialism, cases of South Korea and other
Asian Tigers

3. Adversarial legal system (India), Investigative judiciary (USA), common law vs civil law systems, parliamentary vs presidential
forms of government, insolvency and bankruptcy regulation

4. Social norms (e.g. UN Diplomatic Parking tickets and traffic violations in New York City, belief in usurping public goods for private
gain – e.g. socio-cultural beliefs that bribing is ok, dumping paddy crops on public roads for thrashing believing it’s ok), Civil rights
and individual freedoms in China, Syria etc. religion and economic outcomes, festivals, caste system (e.g. Khap Panchayat system in
Haryana, casteism in UP/Bihar/Tamil Nadu)

5. conflict/insurgency/militancy/ terrorism in Syria, Iraq, Sudan, Maoist infested regions in India,

Kleptocracy in Brazil (ex-President Lula), corruption of the Atul Gupta family empire in South Africa during President Zuma’s period,
kleptocracy in Indonesia (ex-President Suharto), oligarchy in Russia (President Putin’s syndicate), cronyism and corruption in poor
African countries
There are 3 types of institutions
• Coordinating institutions
• Extractive institutions
• Productive institutions
Coordinating institutions

Rule: Drive on the left side of the road

Enforcing Organization: Traffic police
Belief or Norm: Drivers will follow the rule.
Regularity of behaviour: Drivers drive on the left side of the road.
Extractive institutions

Rule: Culprits will be punished for their crime and police require at least Rs 10000 to dismiss minor
offense charges (the assumed cost of disposal of the case).
Enforcing Organization: Police, Courts
Belief or Norm: Police find bribes profitable and bribes are cheaper than facing the charges.
Regularity of behaviour: Corruption

Extractive means exploitative. Extractive institutions are much more prevalent in poor, low-income and
developing countries and less prevalent in developed countries that have relatively more robust political and
economic institutions in place.
Productive institutions
Rule: Secure property rights assure investors get return to investment.
Enforcing Organization: Regulatory authority, Courts
Belief or Norm: Financial contract will be enforced.
Regularity of behavior: Investment in enterprise.

Productive institutions and coordinating institutions promote long run economic growth whereas the
extractive institutions reduce incentives to undertake economic activities efficiently and drive down
output growth.
 Good versus bad institutions
• Often, determining what counts as a “bad” institution from the perspective of increasing living
standards can be surprisingly complicated.

• E.g. many countries, including the United States, first became democratic and then achieved
sustained economic prosperity. But many countries, such as South Korea, first became wealthy and
then became democratic. And still other countries, such as Saudi Arabia, became wealthy without
democracy at all!
 Some examples of incidence of corruption in India

• Economy-wide examples:
• Bribery experiences: ipaidabribe.com
• Criminal lawmakers
• Fake news: altnews.in
• Slow judiciary and judges’ corruption
• Police crime (crime committed by police)
• Medical malpractice and graft in health sector
• Electricity theft
• Corruption in Indian industry
• Unethical pricing practices
• Anti-competitive practices
• Rent-seeking

https://scroll.in/article/916022/womens-groups-win-as-supreme-court-cancels-anganwadi-contracts-worth-rs-6300-crore-in-Maharashtra
 Size of India’s shadow economy as % of GDP 1991-2015

31
29
28.43 28.02
27.96 27.83
27 26.526.67 27.07
26.96 26.726.62
26.48
25.69
25 24.84
23.87
23.44
23
22.27
22.06 21.68
21 21.03 20.65
19.71
19 18.99
18.33
18.11
17.89
17
15
1991
1992
1993

1995
1996
1997
1998
1999
2000
2001
2002
2003
2004

2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
1994

2005
Data Source: Leandro Medina and Friedrich Schneider, 2018. “Shadow Economies Around the World: What Did We Learn Over the Last 20 Years?” IMF Working Paper WP/18/17.
 Does corruption pay off? Competing views on corruption

• Corruption increases growth as it “greases

the wheels” of economic growth, i.e. it
increases bureaucratic speed and effciency.

• Corruption decreases or ‘sands the wheels’

of growth as it distorts incentives, raises
inequality and leads to sub-optimal
outcomes.
 Why is corruption high in India?

• Incentive problem

• Corrosion of institutions (legislature, executive, judiciary, media, corporate sector)

• Misery (inequality, unemployment and poverty)

• Lack of civic engagement

 Sukhtankar and Vaishnav (2015)’s view:
Corruption is high in India because of…
(a) Regulatory complexity
• License Raj, Crony Capitalism
• Low ease of doing business
• Maze of laws and regulations
• Rent-seeking lucrative sectors (mining, land, spectrum, public resources)
(b) Lack of enforcement/ implementation capacity
• Low state capacity, slow judiciary
• 122.5 police officers per 1 lakh population
• 16.5 judges per 1 million residents
• 1 doctor per 11 thousand population

Source: Sandip Sukhtankar and Milan Vaishnav, 2015. Corruption in India: Bridging Research Evidence with Policy Options. India Policy Forum.
 Corruption is high in India because…

(c) Inadequate regulation of political

finance

(d) Public sector recruitment, postings, and

transfers

Source: Sandip Sukhtankar and Milan Vaishnav, 2015. Corruption in India: Bridging Research Evidence with Policy Options. India Policy Forum.
 Kaushik Basu’s proposal to legalize bribe-giving, not bribe
taking

• Harassment bribes are the cash/kind demands from officials to do what they are required to do, i.e. grant
what the prospective beneficiary is entitled to.

• On decriminalising those who pay such 'harassment bribes’, bribe-givers will have more incentive to give
evidence against corrupt officials. Because both are currently criminalized by the offer of a bribe, each
has an interest in keeping quiet about it.

• Giving complete immunity to the bribe-giver would ensure higher reporting and cooperation of the giver
in bringing to justice the bribe taker.

• If he/she could foresee such a situation, the official would not ask for bribe. The fear that a bribe giver
may report the public official could reduce corruption, at least in terms of harassment bribes.

Source: Kaushik Basu’s (2011) note “Why, for a Class of Bribes, the Act of Giving a Bribe should be Treated as Legal.”
Various criticisms

• Moral issues: This proposal may reduce the stigma attached to bribe-giving and result in corrosion of
morality.

• Details matter: Related literature on tax amnesties, leniency and whistle-blowers suggest these
schemes are subtle: effective if well designed and administered, counterproductive if details are not
set right.

• Power imbalance: Basu’s proposal overlooks the power imbalance between the usually poor bribe
givers and the powerful officials who demand the bribes. Those giving may be too afraid to make
notes or take secret photographs to entrap officials.

• Scope for vendetta: If you refused to pay, or entrapped an official, other officials could take action
against you as a vendetta/ revenge because of their collusion with their colleague.
Jean Dreze’s criticisms

1. What if law enforcement agency (police, judiciary) is also inefficient and corrupt as the rest of the
bureaucracy? Reporting may bring “litigation costs, possible harassment and little chance of
getting justice”.

2. If there is a positive probability of being convicted for bribing in the absence of reporting for
bribery, then legalizing the act of bribe-giving may induce people to pay bribes that would
otherwise be deterred.

3. Legalizing bribe-giving might reduce the moral cost of paying bribes.