Development economics - Theories

Banerjee and Duflo

Essay about the life of the poorest people in the world

Banerjee and Duflo - Extremely poor

Living on less than PPP $1.08 (today $1.90)

• in paper 30% of global pop

• Today 8% due to China eradicating poverty

• OBS.: find more specific why this definition

Banerjee and Duflo - Poor

living on less than PPP $2.16 (today $3.20)

• in paper 50% of global pop

• OBS.: find more specific why this definition

Banerjee and Duflo - Food consumption (check it is them)

Universal pattern

• Food represents from 50-75% of total expenditure (TE)

• alcohol and tobacco from 1-8%

• a 1% increase in E means a 0.67% increase in food consumption

• 1% increase in FE is equally located towards buying more calories and buying more expensive (tastier) calories → same across the world → behavioral

Banerjee and Duflo - Durable goods

Nonuniversal pattern

• Radio, television and bicycle ownership varies significantly from country to country and is low many places due to infrastructure constraints

• steep income gradient in ownership of these assets

Banerjee and Duflo - other entertainment

Weddings, funerals and festivals is around 10% of annual budget

• culture plays a big role

Banerjee and Duflo - Why don’t they eat more/better?

• Eating more or better does not help much. Nutrition explains only a very small part of health gains worldwide over the past few decades (Disproven after). However, evidence of improvements in nutrition (reduction of anemia) is directly linked to increased productivity (Not longer life expectancy but higher quality of life)

• lack of saving for difficult times does not fully explain low food consumption - because poor HHs do spend on entertainment and save up for events, gadgets, etc.

• The need to spend on entertainment rather than on more/better food is probably a network effect

Banerjee and Duflo - How do poor people generate income?

• Many are entrepreneurs

• often involved in multiple occupations (median family has 3 working members and 7 occupations)

• limited landholdings and an incomplete rental market leads to relatively small-scale agricultural production

• non-farm businesses also small scale

• temporary migration is an important source of non-farm income for rural poor. migration is short term and distance

• avg. working hours: 84 per week

Banerjee and Duflo - Why are they entrepreneurs and so many occupations?

• few skills, little capital and there are rigidities and information problems in local labor markets → being small-scale entrepreneur = easier than finding a job

• cannot raise the capital necessary to run full-time businesses (specialize)

• risk spreading difficult because of missing credit instituions and lumpy incomes

• to be occupied: each occupation is limited in time use

Banerjee and Duflo - Credit

• the fraction of rural poor having outstanding debt varies a lot (11-93%)

• very few of the loans from formal lending sources

• credit from informal sources = expensive (>3% per month)

• informal interest rate lower if HH owns land

• frequent delays in payment are very common, but default is not

• low default rater ≠ automatic ← high contract enforcement costs

Banerjee and Duflo - Savings

• few HHs have bank savings and despite active promotion, share of poor saving with semi-formal institutions is still low

• → poor save informally

• financial illiteracy

Banerjee and Duflo - Insurance

• very little access to formal insurance (including health)

• access to well-functioning informal insurance schemes through social networks

  • → only limited protection

Banerjee and Duflo - Assets

• land tends to be the one asset the poor own

• large variation in ownership and plot size

• land records often incomplete → many do not have the titles to their land

Banerjee and Duflo - Health and happiness

obs.: Collins et al?

• avg. poor person consumes less than 1400 calories per dag

• only 57% report HH had enough to eat in the year

• avg. BMI is 17.8 (normal 18.5-25) (now peak at 25-29 for highest life expectancy)

• many complications due to poor health

• 11-46% of HH report having member either bedridden for the day or requiring a doctor in past month

• BUT selfreported happiness not very low. cutting meals is strongly correlated with reported unhappiness

• BUT report high levels of stress - financial and psychological

  • most cited reason = health problems

Approaches to poverty analysis

• Monetary

• Capability

• Participatory

• Social exclusion

Monetary approach to poverty analysis

poverty is a shortfall in consumption (or income) from some poverty line

• used by Banerjee and Duflo

• Ravallion

• Preferred by economists

  • Total expenditure is preferred in developing countries

  • Disposable income is preferred in developed OECD-countries

Three basic steps:

1. Defining one (multiple) indicators of well-being

2. establishing a minimum acceptable standars of that indicator to separate the poor from the non-poor (poverty line)

3. generate af summary statistic to aggregate the information from the distribution of welfare indicator relative to the poverty line

Capability approach to poverty analysis

Poverty is a failure to achieve certain minimal or basic capabilities, where “basic capabilities” are the ability to satisfy certain crucially important functionings up to certain minimally adequate levels

• Sen

Participatory approach to poverty analysis (PPA)

Poverty is local people’s perception of well-being and ill-being

• world bank

Social exclusion approach to poverty analysis

Poverty is a process through which individuals or groups are wholly or partially excluded from full participation in the society in which they live

• Townsend

Theoretical monetary poverty line

Minimum cost of a household of attaining a given level of utility at the prevailing prices and for give household characteristics

• For given characteristics and prices, the HH minimizes their expenditures to get a given utility level: e(p,u|x)

• The poverty line is the E needed to be exactly nonpoor: z(x)=e(p,u_z |x)

measuring expenditure

ALWAYS BASED ON SURVEYS

• otal expenditure is not known to anyone in any country, specifically we rarely know:

  • value of durable goods

  • value of publically provided goods (edu and health)

• Getting poverty measures = collect data from nationally representative HH surveys - country by country and year by year

  • World Bank has organized surveys since 1985: Living standards measurement survey (LSMS)

  • Many countries now organize their own household surveys using LSMS as a frame

The Identification problem - Problems

• poverty comparisons are based on individuals but data is HH level → we need to move from HH to individual level

Problems:

• we know very little about intra-HH allocations

• children typically consume less than adults

• HH have economies of scale in consumption

The Identification problem - Solutions

• Many countries use “equivalence scales”:

  • 1+0.7*(adults-1)+0.5*children

  • sqrt(HH size)

• Most deveveloping countries simply divide total HH expenditure by the number of members of HH and obtain expenditure per capita

The reference problem - Problems

We use consumption expenditure with reference prices as a measure of wellbeing and seek the threshold between poor and non-poor

Problem: Finding the threshold and is it

• Absolute

• Relative

Note: economic theory has very little to say here

• Ravallion argues that the choice depends on the purpose

Absolute poverty line

A fixed level of expenditure based on

• subjective criteria

• objective criteria

Steps:

1. Pick a nutritional requirement for good health (fx 2100 kcal/day)

2. Estimate cost of consuming this energy requirement, using a food consumption bundle that reflects the habit of HH near the nutritional requirement. This is the food poverty line b^f

3. Add a non-food component, bⁿ

4. The overall poverty line is the sum: z=b^f+bⁿ

Relative poverty line

An expenditure level relative to the median or mean level in the pop (characteristics, includes HH size and location)

• welfare may depend in part on relative incomes

  • evidence from subjective self-assesments of welfare

  • may explain the Easterlin paradox

• Ravallion always thinks about poverty comparisons

Absolute poverty line - food part

Countries choose their own “nutritional anchor”

• differenfces due to: culture, genetics(?), how we measure poverty

  • Choose smaller threshold → less poverty → less donors (example Uganda having 3000 kcal/day as threshold)

Absolute poverty line - nonfood part

• Typically determined as a fraction of food expenditure

• specific fraction impossible to determine theoretically, but there are some limits

• often: expenditure shares of the reference group is used to calculate the non-food expenditure

• or: poverty line is found from an engel curve regression

Absolute poverty line - nonfood part - engel curve regression

Assumptions:

1. once survival food needs are satisfied, as total expenditure rises, basic non-food needs will have to be satisfied before basic food needs

2. once survival food needs are satisfied, both food and non-food are normal goods

Outcome: The poverty line cannot exceed the TE of those whose actual food spending achieves basic food needs

Global poverty: History of the “$1 a day”-poverty line - background

• all countries with LSMS type HH surveys have their own poverty line expressed in the national currencies

• PPP numbers for a wide range of countries make comparisons of HH surveys, and thus poverty, possible

• BUT: national poverty lines dont collapse to a single line even when using the PPP-dollar (NOTE: slope is approx. 0.5 → less systematic relationship than earlier paper)

• National poverty lines are increasing in the income/consumption level above some critical level

Global poverty: History of the “$1 a day”-poverty line - How it was decided

• WB proposed a line of $0.76 a day at 1985 consumption PPP based onf 33 national poverty lines for the 70’s and 80’s

  • = the predicted poverty line for the poorest country in the sample

• Higher line of $1.02 a day was more representative of the poverty lines in low-income countries

• →higher line became more accepted in WB and internationally

• In 2001: WB revised past estimates and found:

  • poverty line for poorest country: $1.05

  • more representative: $1.08 at 1993 PPP

Ravallion, Chen and Sangraula (2009)

Estimated new “global poverty lines” using 2005-PPP data and poverty lines from 75 developing countries

• $1.25 a day in 2005-PPP was representative

Latest global poverty line

$1.90 a day using 2011-PPP

• WB (2018)

Followed 3 basic principles:

1. use the most accurate and recent set of prices available to compare the real standards of living across countries

2. minimize changes to the goalposts: keep the definition of the line unchanged, and its new values as close to the $1.25 line in real terms

3. when defining “real terms” the price levels that matter for measuring global poverty are those faced by the world’s poorest people

How: used Ravallion lines for 15 very poor countries → inflate to 2011-prices using each countries CPI → convert to $US using 2011-PPP → simple average

Allen (2017)

Absolute poverty measure based on the minimum cost of nutritional adequacy

• calculate the bundle that minimizes the cost of attaining predetermined nutritional requirements

Pros

• Avoids concerns about how PPPs are constructed and used in global poverty measurement

Cons:

• minimum cost diets are typically very inexpensive, but exceedingly dull and very often regarded as unacceptable (Sen). The least-cost method not used a lot. Methods that identify a food bundle consistent with prevailing tastes in each setting, respecting the influence of local food habits as well as recommended nutritional intakes, used instead

Does it matter which absolute measure we use over time?

NO

• both strengths and weaknesses in WB and Allens’ methods

• the trends over time are very similar, despite the methodological differences (Ravallion, 2020)

The Easterlin paradox

Avg. happiness appears not to rise much with economic growth

Strongly relative poverty lines - setup

Compare HH income to a poverty line that is set at a constant proportion of the current median for the country: z^R=k*y(pi_z)

• z^R is the relative poverty line

• k is a constant

• y(.) is the quantile function

• pi_z is a fixed percentile that defines the comparison group (often the mean or median)

Strongly relative poverty lines - Issues

It is not clear why the quantile of any fixed percentile identifies a plausible comparison income

• when the poverty line i set at a constant proportion of the mean (or median), the resulting poverty measure depends solely on the distribution of relative incomes in the pop

• If all income levels grow at the same rate, then the poverty measure will remain unchanged (as k is constant)

• the poverty lines in developing countries have an avg. elasticity to the mean of about 0.5, which is significantly less than 1

Weakly relative poverty lines - setup

z^R=z^A+beta(m-z^A)

• z^A is the absolute poverty line

• beta is a parameter

  • if 0<beta<1, then the elasticity of the poverty line wrt. m is positive but less than unity

• m is the mean or median

  • if m<z^A:

    • z^R=max(z^A,alpha+beta*m)=z^A+max(alpha+beta*m-z^A,0)

    • alpha can be interpreted as the lower bound to social inclusion needs

Weakly relative poverty lines - Empirical

• z^A = $1.90 a day in 2011-PPP

• weakly relative lines calibrated to national lines = Rank-weighted mean is the relevant comparison income, with lowest weight given to the richest

• Implies that a Gini-discounted mean should be used: m*=(1-G)m, G=Gini index in each country

• z_j^(A+R)=max($1.90,0.9+0.7(1-G_j)m_j)

  • why alpha=0.9 and beta=0.7

Implication: A person is not poor globally if she is neither absolutely poor (relative to z^A) nor poor by the expected standard for the country she lives in

Absoulte vs. weakly relative poverty measure

• over 90% of the poor by the A+R line are found in the developing world

• with the falling number of absolutely poor in the developing world, we see rising relative poverty

• Conclusion: progress in reducing global poverty is evident. but we observe rising counts of those who are relatively poor but no longer absolutely poor. what is the correct SDG measure of poverty?

3 core values of development

• Selfesteem: to be a person

• Freedom from servitude: to be able to choose

• Sustenance: the ability to meet basic needs

3 objectives of development

• increase availability of life-sustaining goods

• raise levels of living

• expand range of economic and social choices

Important capabilities

• being able to live long (health → income?)

• being well-nourished (nutrition, health → income?)

• being healthy (nutrition, health, lifestyle → income?)

• being literate (education → income?)

• being well-clothed (income)

• being mobile

• being able to partake in the life of the community (politics)

• being happy - as a state of being - may be valued as a functioning (?)

Instrumental roles of freedom

• some freedoms are not part of the core values, but they are instrumental

• instrumental freedoms directly enhance the capabilities of people, but they also supplement and can reinforce one another

5 important instrumental freedoms:

1. Political freedoms

2. economic facilities

3. social opportunities

4. transparency guarentees

5. protective security

2030 Agenda for sustainable development

Adopted by all UN member states in 2015

• has the 17 sustainable development goals (SDG)

GDP as a measure for development

• Close link between economic development and economic growth, but not same clear link to quality of life

• GDP growth is not a sufficient condition for improving mass living standards

  1. leaders may use the added funds on themselves - in a broad sense

  2. the gains from growth may be invested to gain furhter growth

  3. the gain may accrue income only to the already well-off

GDP as a measure for development - measurement issues

• Non-market is not (fully) counted in GDP

  • if home production relative to the market exchange declines during development then “economic wellbeing” is underestimated by GDP

• Unreported income from selfemployment is not (fully) counted in GDP

  • if the informal sector is relatively larger in poorer countries then the “economic wellbeing” is under estimated by GDP

GDP as a measure for development - Conceptual issues

• GDP does not take the main “freedoms” into account

  • political, social, transparency guarentees, protective security

• GDP may decline if leisure expands (reduced market participation)

• GDP can expand through increasion resource extraction

• Market prices may not be the right prices (externalities)

• GDP has no information about the distribution of income and wealth

Purchasing power parity (PPP)

• The number of currency units required to purchase the amount of goods and services equivalent to what can be bought with one unit of the currency of the base country (Usually $US)

• needs: prices of comparable and representative items and expenditure weights

The International Comparison Program (ICP)

Two main objectives

1. produce PPPs and comparable price level indexes (PLIs) for participating economies

2. convert volume and per capita measures of gross domestic product (GDP) and its expenditure components into a common currency using PPPs

PPPs are calculated based on the price of a common basket of goods and services in each country

economic development

economic growth

MDGs

HDI

Human development index

Washington Consensus

Market liberalization will lead to higher rates of growth and lower poverty. Reforms might impose short-term pain, but in the long term reformers would achieve sustained growth and development

2 central prescriptions:

• maintain a secure and stable macroeconomic regime by keeping the exchange rate competitive, the budget deficit low, the tax base broad

• Improve markets by liberalizing foreign trade, abolishing restrictions on inward foreign direct investment, eliminating preferential interest rates, privatizing public enterprises and protecting private property

Washington Consensus - the Wade view

three main problems:

• Prescriptions did not emerge out of a record of what had been shown to work

  • prescriptions based on “basic” economic theory (Neoclassical economics)

• Focused on the “market”

  • prescriptions ignore the task of building the capacity of industries and firms, including the capacity to compete in international trade. Assumption: No “serious” market failures can be solved by state intervention

• Prescriptions are couched as valid for countries at all stages of development

  • ignore the negative “late development effect”: countries which begin to industrialize when other countries are already highly developed have to use different policies and institutional arrangements to those used by earlier developers to be “competitive” (POWER)

  • critique assumes that this negative late development effect dominates advantages of coming late (such as better and cheaper access to more advanced technologies)

Washington consensus - Background

• coming from a perceived failure of “Third world developmentalism”

  • Infant-Industry protection strategy

• Two forces led to the demise of “Third world developmentalism”

  • radical change in the “leading” heoretical approaches within econoics. move away from the postwar keynesian consensus toward orthodox neoclassical ideas

  • realignment in the global political economy diminished the bargaining power if third world governments

Washington Consensus - Politics

Coordinated campaign of international bureaucracies exerting their newfound power to promote radical international political changes

• US was the leading engineer

• Baker Plan

• WB publications moved away from discussions of poverty and toward discourse around the virtues of market liberalization

The Baker plan

Financial aid to indebted governments would be used to incentivize market-liberalizing reforms. the main vehicle for achieving this was conditionality

Washington consensus - in practice

• WB reorganised lending around comprehensive plans of reform → policy conditional “structural costs” loans

• speed of reform was the most important

  • try to reach a PONR so the policies were self-enforcing

• Goal was more political alignment than development

Washington consensus - The actual white paper

• never intended as a fully elaborated plan, growth strategy or a model of development

• not a statement of necessary or sufficient conditions for growth and development

  • One size fits all has neveer been supported academically

Common elements of sustained growth experiences

1. macroeconomic stability

2. high levels of investments and saving

3. good leadership and governance

4. success in exploiting the international market for technology/knowledge (facilitating environments for spillovers)

5. diversified outward orientation - advancing economic complexity

6. allowing the market to play a role in resource allocation

7. managing the distributional aspects of growth patterns to put boundaries on inequality in various dimensions

Washington consensus - Missing elements

• underemphasized the role of knowledge transfers

• structural change is a key element in growth dynamics NOT side effect

• Relying on private sector solely will not lead to growth as public investments are crucial

East-Asian development strategy

• government played a larger role

• gradual “fence-breaking” and a lot of experimentation

• free market policies were instruments not the objective

inequality measure - principles

1. the anonymity principle: it does not matter who is earning the money

2. the population principle: size doesnt matter

3. the relative income principle or “Scale independece axiom” (SIA): size really doesnt matter

4. the transfer principle (Pigou-Dalton): a transfer from a poorer to a richer person will increase the inequality

If all are followed the metric is lorenz consistent

The translation invariance principle

• replaces SIA

• give everyone the same amount of money and inequality is unchanged

• but: the inequality measurement is affected by increasing all incomes by X%

The Kuznets ratio

Used to compare lorenz-curves

income of x% richest/Income of y% poorest

• does not include the whole distribution

• fails to satisfy Pigou-Dalton

The gini coefficient

• area between the 45-line and lorenz curve

  • includes the whole distribution

  • satisfies the four fundamental principles: lorenz consistent

Theils indexes

Mean-log-deviation

• Includes the whole distribution

• lorenz consistent

• used by Milanovic

• can be divided into subpopulation

Coefficient of variation

(standard deviation)/mean

• includes the whole distribution

• lorenz consistent

Gini coefficient - issues

• information is lost going from the distribution to the single measure

• intersecting lorenz curves can cause difficulties if only looking at the gini coefficients. Lorenz consistent summary statistics can give conflicting results when the curves cross

• a decreasing gini ≠benefit to the poor

• it isnt decomposable

• the absolute dimension → equally poor is not a desirable outcome

Inequality within countries - Trends

• In US inequality has risen by Gini and share of the top decile in national income

• Top 1% share of total income has a U-shape in US, and L-shape in Europe and Japan

• ROW: South asia least unequal, Namibia very unequal

Milanovic - Cross-country and global inequality - measures

• Concept 1: Cross-country gini based on income per capita across countries without weighting

• concept 2: Cross-country gini based on income per capita across countries with weighting

• Concept 3: individual incomes

Milanovic - Cross-country and global inequality - empirical

• about 8% of the world receives 50% of global income

• global inequality showed no clear trend, despite the development in China and India

• Almost certainly the highest level of relative and absolute inequality at any point in human history

Decomposition of global inequality - individual incomes

• slight increase in inequality within countries

• decrease in inequality between countries

• net effect: decrease in global inequality

The elephant curve

• Lakner and Milanovic

• % real income change X percentile of global income distribution

• Losers: the very poor, sub-sharan Africa - incomes are unchanged

• Winners: middle classes: China and india - incomes risen substantially

• “start of trunk”: citizens of rich countries with stagnating wages + pop of former communist countries - big share of income + stagnating income

• “End of trunk”: the very rich - half are the top 12% of US - big share of income distribution + rising income

If looking at absolute gain in income:

• steep curve at the richest part → sharp increase in absolute inequality

Inequality - Class or location

• Composition of global inequality changed from differences within countries to differences between countries

• more than ¾ of global inequality is due to between country differences

• Milanovic: Impact on our political systems

From growth (development) to inequality

• Kuznets inverted U-hypothesis

  • uneven and compensatory changes

• Education and skills (biased technological skills)

• wealth accumulation

• seeking support for growth-oriented development strategies

From inequality to growth (development)

• physical capital accumulation:

  • Classical theory (Kaldor)

  • Richer families have higher savings rates. Therefore, redistribution in favor of the rich increases savings and thereby accumulation of physical capital

• human capital accumulation:

  • Galor and Zeira

  • Human capital is not transferrable. If accumulation of HC is costly and capital market imperfections or restrictions → poorer people will underinvest

• political redistribution:

  • modern political economy model

  • inequality leads to desire for redistribution by the median voter. distortionary taxes → lowers investments and thus growth

• Capital markets

• Inequality augmented growth regressions

• support for equity objectives in development strategies

Kuznets inverted U hypothesis

looked at inequality data across developing and developed countries in 50-60s

• across countries inequality appears to be first rising with avg. income and subsequently falling againg

• As a country developed inequality first rises the fall

  • fx due to: transition from traditional, poor sector (farming) to moderne wealthier sector (industry) - NOT KUZNET THOUGHT

Kuznets inverted U - Empirical

• Find all shapes - Anand and Kanbur

• 90% times no inverted U-shape - Deininger and Squire

• Conditional Kuznet-curve - Barro

The conditional Kuznet curve

Conditional on:

• History and politics (Apartheid i SA vs. land reform in South Korea → one increased the other decreased inequality)

• Resource endowments

  • Mineral wealth

  • Farming systems (rice vs. plantations)

• Contemporary policies

Ravallion (2005)

Are we seeing falling poverty with rising inequality, and rising poverty with falling inequality?

• Only considers absolute poverty

Conclusion:

• no trade-off between absolute poverty and relative inequality

  • positive correlation

  • growth splits - does not change results

• trade-off between absolute poverty and absolute inequality

  • negative correlation

  • growth splits - some indication of differences

If relative inequality does not change on average with growth then absolute inequality will tend to rise with growth

The Ravallion “Trick”

Is there an unconditional correlation between poverty and (absolute and relative) inequality?

• trick: study the changes over time in both measures of poverty and inequality

  • assumption: controlling for “fixed effects” will adequately sweep away any extraneous other factors at country level

• Measures:

  • relative inequality: Gini

    • % change in poverty=beta*(% change in Gini)

  • absolute inequality: gini controlling for mean income or consumption

    • % change in poverty=beta*(% change in Gini + % change in mean income)

Growth accounting

The relative importance of factor accumulation and productivity growth

• compare the country to itself!

1. how much of the variation in growth rates among countries is explained by variation in produtivity growth, and how much by variation in factor accumulation

   • look at how much of th growth is caused by k, h and A

2. How much does productivity growth differ among countries

Development accounting

A decomposition of relative production levels

• compare the country to rich country (US)

Look at the ratio between each of the “growth factors” → are they better or worse

2. How much of the variation in output per worker among countries is “explained” by productivity differences?

   1. How much would output per person change in response to variation in: Phys, human, residual TFP holding the other two factors fixed

   2. how much would output per person change in response to variation in: Capital-output ration, HC, residual TFP holding the two other factors fixed

Solow convergence

1. all else equal, poor countries have the potential to grow faster than rich ones

2. as countries become richer, growth rates tend to slow

3. as consequence of 1 and 2 → income levels of rich and poor countries should converge - all else equal

Convergence debate - Old

• We do not observe unconditional convergence, and there is no reason to expect this

• the steady state in the solow model varies with:

  • savings rate

  • depreciation rate

  • population growth

  • technology growth

  • level of human capital

Convergence debate - New

Three new central facts about developing country growth:

1. since the mid-90s developing countries began to converge toward advanced countries income level. Avg. developing country will close half the gap in 170 years (Patel)

2. coss-sectional growth in the period of convergence exhibits an inverted U-shape. Middle- and low income countries grow faster than advanced since 90’s, middle income countries have been growing faster than all since the 80’s

3. the era of unconditional convergence is also associated with lower volatility and higher growth persistence within developing countries

Patel et al (2021)

Used a convergence equation derived by Barro and Sala-i-Martin

• estimates the speed of adjustment to the SS

• looks at it from t=1960, t+s=2010

• varying number of countries

when using 2000-2019 and 124 countries → avg. expected to have closed gap in 170 years

• Asia drives a lot of the convergence

• without africa it would converge a lot faster

Hsieh and Klenow (2010)

All research on income differences can be classified into one or more arrows in the chain of causality:

Geography, climate, luck → Human capital, physical capital, productivity → income

||
V

Institutions, culture →Human capital, physical capital, productivity → income

||
V

Policies, rule of law, corruption→Human capital, physical capital, productivity →income

Growth accounting - Empirical results

• capital accumulation tends to contribute substantially to growth for low income countries

• increases in the size and quality of the labour force also contributes

• productivity growth tends to account for a larger share of growth in high income countries

Development accounting - Empirical

if alpha=1/3 then:

• 20% of variation in GDP per worker is due to variation in capital-output ratios

• 10-30% is due to variation in human capital

• 50% is due to variation in productivity

Population growth - History

• population growth close to 0 in 1800’s

• introduction of settled agricultural revolution changed the picture despite famines, plagues and wars

• industrial revolution

• annual population growth 1800-1945: 0.6%

• demograph transition

The demographic transition

1. both high birth and death rates

2. death rates declining faster than birth

3. birth rates declining faster than deat

4. zero population growth (again)

TODAY: almost all population growth in high-income countries is attributable to immigration

Population growth - future

• Today: considerable variation in fertility across regions and countries remain

• Will total fertility rates (TFR) stay above replacement rates? Three answers for three reasons:

  1. Desire for large families

  2. unwanted children hypthesis

  3. population momentum

Determinants of mortality

What caused the mortality transition?

• nutritional improvements

• improvements in sanitation

• technology (germ theory early on)

The relative importance of each is in debate: McKeown and Fogel (nutrition) vs Preston (Tech)

Determinants of fertility - population pessimists

• Lower fertility → less investments needed to provide a constant amount of capital per worker (capital widening) or investments used to increase capital per worker (capital deepening)

• lower fertility → public funds can be redirected away from health and education toward physical capital

• lower fertility → slower population growth → lower dependeny ratio → reduction in consumption and increased savings → higher capital accumulation (demographic dividend)

What caused the fertility transition?

three important theories

• Income and substitution:

  • low level of income → income effect dominates, the relative cost of time with children is low

  • high level of income → substitution effect, the children costs too much time compared to the value of work

• Mortality:

  • more children to reduce risk of them all dying

  • as mortality decreased → risk decreased → the no. of children to have for the E(living children)>1 decreased

• Quantity/Quality:

  • accelerating technological changes increases return to skill investments. Better children > many children

• Reinforcing mechanisms: Norms, gender roles, child labor