#economicgrowth

Geography, Climate, Natural Resources & Growth

Summary

Assumptions

Types of Natural Resources

  1. Nonrenewable resources - Changes in level of resources 1. Discovery of new resources 2. Depletion of resources 3. Change in price making new sources viable 4. changes in technology
  2. Renewable resources ΔS_t=S_t+1-S_t=G_t-H_t where S_t is stock and G_t is growth and H_t is harvested

Model/Theory

Geography that matters:

  • Proximity to the ocean - affects openness
  • Economic spillover effects/clustering
  • Size of states & conduct of government 
    • Unification e.g. China means one large common market, less fighting, faster spread of ideas
    • Decentralization e.g. Europe 1. external competition (check on power) 2. f.o.p. (esp people) can move country. Europe has fragmented areas of fertile regions, India however is a counter-example.
  • Climate - latitude 
    • Agricultural productivity - relating to optimal temperate climate.
    • Diseases is more common in tropical regions.
    • People work harder in colder climates.
  • Natural resources
    • Resource curse
      1. Overconsumption (temporary booms) --> lower savings to maintain same level of consumption
      2. Dynamics of industrialization (shift away from manufacturing domestically - low transports costs and production abroad means there are less and less backward and forward linkages
      3. Politics - government spend the money on investment/infrastructure/education? - Often the presence of natural resources --> worst government policies

Global view - resource constrained?

  • Demand & Supply usual keeps scarcity of resources in check through higher prices but the environmental damage is largely an unpriced externality.
  • Renewable resources lead to a hill-shaped yield curve (p.488)
  • Lack of property rights can lead to inefficiently high overuse (tragedy of the commons) - biggest concerns are those that aren't owned by even governments - seas and air.
  • With I=R/yL where L = population, I = resource intensity R = resource consumption and thus R-hat=I-hat+y-hat+L-hat.
  • Can rework GDP accounts to include 'natural capital' and therefore account for the limited natural resources, were resource prices are adjusted to account for future externalities (q: are prices correlated with scarcity?)
  • Economics needs scarcity to be reflected in factor prices which it currently isn't.
  • Environmental Kuznets curve argue that initial development worsens the environment but in the long-run its better. Could potentially be gains from trade re: pollution
  • Global Warming - simple solution carbon tax.

Predictions

Evidence

  • There is a strong positive correlation between distance from the equator and gdp/capita, where although countries near the equator can be poor, countries far from the equator never are.
  • Positive correlation between natural capital and GDP/capita. Growth rates are negatively correlated with natural capital that makes up large proportion of the share of national wealth.
  • If oil reserves are estimated at 3.0 trillion barrels we have already used 39% and at that rate all the oil will be gone in 61 years. And if usage continues to grow at 1.6% a year (the average rate from 1983 to 2010) all the oil will be gone in 43 years

Evaluation

  • Geography/natural resources do not suffer from reverse causality.
  • Natural resources are not essential for economic growth and they may actually be a hinderance.
  • Resource constraints can be overcome by 1. substitution (SR this is hard, in the long run substitutes are found) and 2. technological progress

Culture & Growth

Summary

Assumptions

What determines culture?

  1. Religion
  2. Climate & Natural Resources
  3. Cultural homogeneity and social capital - homogeneity helps growth by social capital and stronger networks.
  4. Population density & social capability
  5. Government policy
  6. Income/capita & the economic environment
  7. Ethnic make-up
  8. Media/TV

Theory/Model

  • Ways culture matters: 
    1. Openness to new ideas
    2. Belief in the value of hard work
    3. Saving for the future
    4. Mutual trust - f(state, reputation).
      • Social capital - large social networks keep people accountable and higher trust feeds through to higher income/capita
    5. Social capability
      • peoples' experience of large-scale organisations
      • ability of residents to take advantage of market economics/specialization/trade
      • cause & effect over superstition or magic
      • life on earth > spiritual existence
  • Culture & Income/capita
    • Exogenous change in economic environment first leads to a shift B and then over time with a change in culture a shift to C.
    • This can lead to a positive multiplier effect.

 

Predictions

Evidence

  • Interestingly there is a negative relationship between belief in the value of hard work (versus leisure) and income/capita. However, it would be interesting to know if culture (with similar income/capita) is a good predictor of economic growth. Japanese in 1900 were deemed lazy and content. 
  • Studying immigrants to new countries found there was no correlation between the saving rate of an immigrant and the saving rate of the country they came from i.e. no relationship between culture and saving. Although immigrants are not a randomly selected group.
  • Law abidance has a cultural component (see p425 for parking tickets and culture).
  • Strong positive relationship between social capability and future economic growth (1960-2009)
  • A negative correlation between ethnic fractionalisation and gdp/capita, homogeneity is good - obviously may not be causal.
  • Strong relationship between population density in 1960 and growth rate in gdp/capita
  • Telenovas in Brazil were a significant contributor to the rapid reduction in Brazil's total fertility rate, from 5.8 children per woman in 1970 to 2.9 in 1991. 

Evaluation

  • Beware of observer bias - a country is successful therefore it must be because of its culture.
  • Cultural characteristics may be good at some times for growth and bad at other. 
  • How persistent are cultural traits? Do they last over multiple generations?

Government and Growth

Summary

Assumptions

Model/Theory

  • Government should solve market failures 1. public goods. 2. externalities 3. monopoly 4. coordination failure and 5. income distribution.
  • Government role: 
    1. Rule of Law
    2. Overall size of government
      • Tax distortions. Laffer curve.
    3. Practice of planning
      • Historically included state-owned enterprises, government-owned banks, marketing boards, trade restrictions.
  • Why do governments do things that are bad for growth?
    • Other goals e.g. reduction of pollution, equity-efficiency trade-off (redistribution).
    • Corruption & Kleptocracy --> 1. corrupt govs spend more (not to lowest bidder) 2. policies pursued purely as opportunities for bribery e.g. quotas 3. undermine rule of law
    • Self-preservation - economic growth can be destabilizing.
  • Is the causation bad government --> poverty? Or poverty to bad government?
    • Income --> gov 1. bad government not always an impediment to growth 2. quality of gov improves with growth 3. can pay civil servants high wages 4. larger pie to share 5. honest gov. is a luxury good only rich countries can afford
    • Gov --> Income 1. Evidence gov can affect economy 2. Colonialism led to bad govs but was not due to income  - weak institutions and poorly constructed national boundaries (concerning ethnic groups etc.)
  • Income inequality
    • Kuznet's hypothesis - as a country develops inequality would at first rise than fall. 
    • inequality --> higher growth
    • Causes - Distribution of economic characteristics and circumstance e.g. technological advances, increases in international trade, superstar dynamic
    • Effects of inequality 
      1. Physical capital accumulation - more saving if more rich people (better initially for growth)
      2. Human capital accumulation - more education if more equality (better in the long-run for growth)
      3. Productivity = technology & efficiency e.g. inequality --> redistribution via taxes --> inefficiency from distortions and tax avoidance; there may also be pressure for redistribution from crime political instability
    • Economic mobility e.g. intergenerational mobility --> 1. better use of human talent in society 2. less political pressure for redistribution. causes of mobility 1. education access 2. nature of institutions and government 3. nature of marriages

Predictions

Evidence

  • Strong relationship between rule of law and factors of production (p.358) and rule of law and productivity (p. 359)
  • Wagner's Law says that the size of government increases as countries become wealtheir because a more developed economy requires more complex regulation and because many public goods require spending that rises more than proportionally with income (p.360)
  • Robert Barro (p.377) found some democracy is good for growth but too much is bad.
  • Average income level rather than level of inequality is the key factor in determining an individual's income level.
  • No statistical relationship between inequality and instability, or inequality and higher levels of redistributive taxation.

Evaluation

  • Reasons against government intervention 1. government failure 2. can successful privatizse many public goods 3. can successful deregulate many monopolies 4. equity-efficiency trade-off wrong where most redistribution is life-cycle related rather than real redistribution - and potentially negative efficiency costs.
  • Swings in opinion in the 1920s with the success of the Soviet Union and the subsequent Great Depression economists were in favour of government intervention, since WW2 and the failure of Communist countries economists are more laissez-faire but once again since the 2008 recession the pendulum is potentially swinging back the other way as China faired better than say the USA.
  • Economic effects of government are sometimes a result of the fight for power rather than economic policy per se. E.g. Paul Collier has cited the multiple equilibria 'conflict trap' or a peaceful & prosperous country.

Free Capital Flow Model

Summary

Assumptions

  • Based on Solow Model
  • Economy is open to capital flows therefore use law of one price where rental rate is the same globally.
  • Small economy (compared to rest of the world) - i.e. no effect on global factor prices
  • Ignore human capital
  • r=MPK=αAk^(α-1)
  • r_w=r

Model/Theory

  • k=(αA/r_w)^1/(1-α) This implies that capital/labour ratio depend on world rental rate of capital - whereas in previous Solow model only domestic factors like the savings rate and growth rate of population mattered.
  • y=Ak^α=A^1/(1-α)*(α/r_w)^α/(1-α). This implies that saving rate is irrelevant for level of GDP/capita!

Predictions

  • For countries with high savings rates openness to capital flow will lower the level of GDP/worker as capital will flow abroad to countries where its MPK is higher (because the increase in capital stock would lower the MPK domestically). However GNP is still higher in both high and low saving countries when open vs closed.

Evidence

  • If free capital model holds savings and investment should be uncorrelated whereas in a closed model they are perfectly correlated. Having 1960-74 study found saving and investment are highly correlated therefore the assumption of free capital movement is inappropriate. Savings retention rate had fallen from 0.89 in 1970s to 0.60 between 1990-97.

Evaluation

  • Crucially a high savings rate may still make a country better off based on GNP/capita and owning overseas assets.
  • For a country with a low savings rate openness should raise GDP, at least much more rapidly than it could from domestic savings.
  • In general openness to free capital is not sufficient to be the primary channel of increasing GDP/capita - thus openness must improve productivity.

Technology: 2-Country Model

Summary

R&D spending in a country will have two effects 1. change the country's relative position in the world technology hierarchy, with transitory growth in both technology and income 2. increased R&D will lead to faster growth in technology for the world as a whole.

Assumptions

  • Technology is nonrival and frequently nonexcludable
  • One-country model key equations:
    1. γ_A=L_A/L (fraction of labour force in R&D)
    2. y=A(1-γ_A) from L_Y=(1-γ_A)L & Y=AL_Y
    3. A-hat=L_A/μ - where μ is price of the new invention.
    4. If γ_A is constant then γ-hat = A-hat i.e. technology
  • Two-country model key equations
    • wo-ways to develop technology 1. innovation 2. imitation, cost of latter (μ) is lower.
    • Assume technology leader has higher A, fraction of labour force in R&D (γ_A,1>γ_A,2) than technology follower but same size labour forces - collectively guaranteeing one is technology leader.
    • μ_i>μ_c - where cost of copying decreases with a larger gap in technology.

Model/Theory

One-country model

  • Increase in γ_A leads to long-run (permanent) increase in A-hat and y-hat but a temporary decrease in y

Two-country model

  • In the steady-state the countries grow at the same rate
  • μ_c=γ_A,2/γ_A,1*μ_i

Predictions

  • One-country model predicts that a country with a larger γ_A will have a larger growth rate. Also larger countries will have more R&D workers therefore larger A-hat.
  • Two-country model predicts that increase in γ_A,1 shift imitator curve up and shift steady-state to the left (smoother curve than one-country model) but crucially there is never convergence in the long-run.
  • Tacit knowledge means 1. much more difficult to transfer between countries than within a country because tacit knowledge is a general rather than specific 2. successful transfer of one technology can lead to large positive externality effects (e.g. taiwan/south korea)

Evidence

  • There is no evidence that larger countries (with larger numbers of researchers) growth faster.
  • Historically there has been a massive increase in the number of researchers but with tech progress being relatively stable.

Evaluation

  • Unlike Solow model increases in capital or savings, increases in γ_A lead to long-term (not temporary) increases in growth rate.
  • Technology leader is not necessarily better off (with a higher quality of living) because it also has to forgo a larger proportion of its workforce to research.
  • Now technological superiority is much more diffuse (unlike perhaps USA post-WW2 and UK in the early 19th century)
  • However technological transfer often doesn't happen for two reasons developing countries 1. choose not to use inappropriate technology (page 246 graph for capital-biased tech change) 2. are unable to use advanced tech (lack tacit knowledge)
  • Embodied technological progress is where new tech is fundamentally tied to capital investment (e.g. new machine). Can lead to leap-frogging where only those countries with the worst tech upgrade, but when they do they go to the best e.g. mobiles>land-lines
  • Assume that cost of technology is independent of level of technology however 1. -ve fishing out-effect of easy ideas 2. -ve time it takes to get to frontiers of knowledge 3. -v marginal value of additional researcher falls with larger numbers of researchers 4. +ve benefits of more areas to work on
  • In the long-run technological growth will probably decline because 1. limited number of new researchers from developing countries 2. fraction of labour force in research cannot grow 3. labour force will not grow. 
  • Tech growth could be zero in long-run if demand is for complements and therefore doesn't increase and f.o.p are shifted to other activities (manufacturing --> services), or if they substitutes it can lead to continued growth (improvements in services)

Productivity

Summary

Assumptions

  • Output=productivity x factors of production i.e. y=A x k^αh^(1-α)
  • Growth rate of output = growth rate of productivity + growth rate of factors of production.
  • Productivity = f(technology, cutting edge of technology, efficiency, open economy)
  • Technology - unlike capital - is nonrival which means easy transfer, but also means there are less incentives for creation because of a lack of excludability.
  • Efficiency = effectiveness with which factors of production and technology are combined to produce output - it's an umbrella concept for differences other than technology where A=TxE

Model/Theory

Productivity

  • Development accounting involves breaking down differences in income into a productivity part and a factor of production part: 
    • Ratio of productivity = Ratio of output / Ratio of factors of production
  • Growth accounting  A-hat = y-hat - αk-hat-(1-α)h-hat

Technology

  • Technology growth can overcome problems of diminishing returns
  • 75% of R&D is done by firms who are motivated by profits and the hope of  1. competitive adv (patents?) 2. size of the market 3. how long will adv last? 4. limited uncertainty

Efficiency

  • Break even point for technology gaps and efficiency is more than 100 years. Thus most of the difference is due to inefficiencies.
  • Featherbedding is when increases in price lead to unions forcing capitalists to pay higher wages/hire workers that are not needed.
  • Types of inefficiency - 1. unproductive activities (e.g. rent-seeking, theft), 2. idle resources 3. misallocation of factors among sectors (due to barriers to mobility &. wages not equal to marginal product - discrimination? 3. misallocation of factors among firms (e.g. government firms/monopoly) 4. technology blocking (not from patents etc.) - often violence related

Openness

  • Two times of international interaction 1. trade 2. flows of factors of production
  • Test for openness is law of one price (though there are transport costs)
  • Globalisation is due to 1. lower transport costs.2. transmission of information 3. more open trade policy (less tariffs, quotas, excessive standards, anti-dumping duties)
  • Openness is highly correlated with GDP/capita but does it cause?
    1. Compare growth rates in open and closed countries & find closed group is 1.5% vs open at 3.1%. Also, closed no relationship between GDP/capita and growth rate, open countries though offer strong evidence of convergence 
    2. Changes in openness affect growth rates? Increased openness e.g. South Korea lead to increased growth. Decreased openness lead to decreased growth.
    3. Used geography as randomly exogenous variable and found that when trade rose, income rose, and when trade fell income fell. 
  • Openness through factor accumulation or productivity? Not factor accumulation as capital flows aren't sufficient.
  • Productivity 1. Comparative Adv/Specialization 2. More able to import existing technologies/capital/business practices. 3. Larger markets (bigger incentives to invest) 4. Increases efficiency by reducing monopoly power/increased foreign competition 5.Economies of scale from increased output

Predictions

Evidence

  • We find large productivity differences between countries where India's productivity per worker is just 31% of USA's. 
  • We find that factors of production (47%) are roughly just as important as productivity (53%) and they tend to rise together. 
  • In US we find that productivity grew at 0.54% per year which makes up 40% (0.54%/1.34%) of the growth rate of output per worker. Across countries we find there is often negative productivity growth in poor poor countries. Across countries we find that 68% of the variation in growth rates is due to variation in productivity growth and 32% is because of factor accumulation. 
  • Alwyn Young found that Hong Kong's growth was primarily productivity, Singapore primarily capital accumulation (esp. human) which is unsustainable. 
  • Russian workers - even post communism are only 20% as productive as US workers - this is a difference in efficiency not technology.
  • Globalisation in two waves 1. peaking pre WWI with world exports at 8% of world GDP 2. now 24% in 2010. Between 1870 and 1925 100m changed countries ~ 10% of the world's population in 1870, much lower today.
  • 1990s study found non-tariff barriers roughly equal to tariff barriers. Though average tariff rates have fallen dramatically from 40% at the end of WW2 to 6% in 2000 and 2.8% among OECD countries.
  • In Japan only 27% of tech progress originated domestically, Canada 3%, US is the only country with a majority being domestic at 82%

Evaluation

  • Problems arise in inadequate measurements of factors of production, making the remainder (productivity) inaccurate.
  • Oftentimes economists overestimate the level of investment (particularly in developing countries) because corruption means much less money is going into companies than they think. 
  • Technological changes comes with creative destruction and potentially harmful side effects for society.
  • Patents have a number of problems incl. 1. inefficiency of monopoly 2. preventing other firms R&D (e.g. delays in patent applications to 3 years) 3. existence of patent trolls. Alternatives incl: 1. secrecy > patents. 2. open source movement
  • Opposition to openness comes from self-interested parties who are going to suffer from trade (though less than net national interest) e.g. a particular industry, or factor of production. Other weak arguments are: 
    • Exploitation - workers wouldn't take those jobs unless they didn't have better alternatives
    • Poor countries can't compete - it's exchange not competition.
    • Environmental exploitation.
    • Loss of national sovereignty
    • Ability to levy taxes (f.o.p. can just leave)
    • Costs of foreign capital 1. subject to international investors --> macro shocks 2. allow go into debt

Human Capital

Summary

Assumptions

  1. Human capital (like physical capital) has five key characteristics 1. it's productive, 2. it's produced (e.g. education) 3. human capital earns a return (but only for working) 4. depreciates
  2. Health and gdp/capita are both endogenous variables of each other.
  3. Human capital is 'installed' in owner whereas physical capital exists independently of owner.

Model/Theory

Human capital can be modelled as:

asefsaefasefsaefsaef.png
  1. Health - people who couldn't work can work and those who could work before can work harder. Exogenous shifts in non-health related GDP/capita or non-GDP/capita related health  shift the curves.
  2. Education - Change Cobb-Douglas production function such that Y=AK^α(hL)^(1-α) where h is related to schooling. Where yss=h[A^(1/1-α)(γ/(n+𝛿))^(α/(1-α)) i.e. steady state income/capita is directly related to labour input per worker (h). Thus comparing countries, ceterus paribus, differences in y will be proportional to differences in h.

 

Predictions

Evidence

  1. Health - Improved nutrition estimated to have a 0.3% effect in UK compared to 1.13 total economic growth. 
  2. Education -  incl. opportunity cost investment in human capital is ~= investment in physical capital (about 12% of GDP in US). College premium (despite declining a bit in the 1970s) has shot up massively - two explanations 1. globalization made education a premium again 2. technological change is skill-biased. Regarding steady state analysis the variations in schooling explains some but not all of the variation in gdp/capita (see page 193).

Evaluation

  • Health - Health view vs Income view: former argues that health is the key problem and if that is solved income will rise, the latter that income is the key problem and if that is solved that health will rise.
  • Education - The reason Cobb-Douglas is better with α of 2/3 rather than 1/3 (which is the observed amount of physical capital) is that there is a return to human capital as well). Other factors that matter include quality of schooling - not just length.
  • Externalities - human capital has positive externalities that physical capital doesn't have - which made lead to an underinvestment on a purely private basis.

Demographic Transition

Summary

Assumptions

Model/Theory

Mortality transition (lower)

  1. Improvements in the standard of living esp. quantity and quality of food.
  2. Improvements in public health measures e.g. draining of mosquito-infected swamps and securing clean water.
  3. Role of medical treatments in curing diseases.

 

Fertility transition (lower)

  1. Improved contraception
  2. Cultural shift in favour of fertility control.
  3. Although total fertility is not that much higher than desired fertility
  4. Economic growth --> lower desired fertility because of  1. lower mortality 2. income & substitution effects (where price of children is higher) esp. opp cost of women not working 3. Resource flows between parents and children 4. Quality-quantity trade-offs.

Replacement rate of fertility is roughly 2.1 children per women would require both an increase in fertility in developed countries and a decrease in fertility in developing countries.

Predictions

  • Predict a generally declining world population growth rate stabilising at around 11 billion by about 2150.

Evidence

  • In developing world mortality rate has fallen faster than fertility thus populations are still growing.

Evaluation

  • Low Total Fertility Rate (TFR) may be due to a tempo effect where women are delaying the age at which they have children however this can only explain 0.25 to 0.4 of the reduction.
  • Demographic momentum is where the number of fertile women boosts population growth even if the rate of fertility stays the same.
  • Demographic structure particularly the percentage of the population of working age is significant for GDP/capita. E.g. In US fraction of working age population is forecast to fall from 0.6 to 0.54 which owould in a 20 year period have a -0.5% effect on GDP/capita.
  • Composition effect means that world GDP is declining as a larger proportion of the worlds population is from poor countries.

Malthusian Model

Summary

Given the right circumstances human can breed at a prodigious rate but are limited by the scarcity of resources - particularly land and by rational 'fertility check'

Assumptions

Model/Theory

An increase in productivity or new land then there would be a shift outwards of top curve but not bottom

Malthus only argued moral restraint could lead to higher standards of living.

Predictions

  • Differences in technology will only result in differences in population size, not standard of living.

Evidence

  • Until recently the model does well for example, in AD 1000, the technologically advanced Chinese lived at roughly the same level of subsistence as the technologically backward Europeans.
  • Model does not apply at all for the last two centuries and in fact the reverse is true, the richest countries in the world have the lowest rates of population growth. Also land was replaced by capital as the key factor of production (and it is not constrained). 

Evaluation

Solow Model

Summary

Capital based model that can explain a lot of, although not all the differences between countries.

Assumptions

  1. Capital has five key characteristics: 1. It is productive 2. It is produced 3. It is limited (rival) 4. It can earn a return 5. It wears out (depreciates)
  2. Production function Y=F(K,L) where we assume constant returns to scale (i.e. F(zK,zL)=zF(K,L) & thus y=f(k) and we assume diminishing marginal product.
  3. Cobb Douglas Production function has an additional assumption where where α is capital's share of income where F(K,L)=AK^αL^(1-α) implies y=Ak^α. In a competitive economy factors of production will be paid their marginal products (i.e. wage=MPL, rent=MPK) and α=MPK*K/Y and 1-α=MPL*L/Y
  4. Constant quantity of labour, L (or at least exogenously determined population & growth rate).
  5. Constant production function i.e. no change in technology A
  6. Capital accumulation is modelled as Δk=i-d where i=γy and d=𝛿k means  Δk=γf(k)-𝛿k

Model/Theory

steady state.png

Steady state can be solved for as γf(k)=𝛿k

  • Kss increases with an increase in γ, or a decrease in 𝛿

ss=A(kss)^α

=A^(1/1-α)( γ/𝛿)^(α/(1-α))

Savings=government savings + private savings

Applications

Predictions

Convergence Towards the Steady State (Solow as a Theory of Relative Growth Rates)

  • If two countries have the same rate of investment but different levels of income, the country with lower income will have higher growth.

  • If two countries have the same level of income but different rates of investment, then the country with a higher rate of investment will have higher growth.

  • A country that raises its level of investment will experience an increase in its rate of income growth.

Evidence

  • Solow Model as a Theory of Income Differences is not very good (Figure 3.7).
  • Solow Model as a Theory of Relative Growth Rates - further away from steady state the fastest a country will grow.

Evaluation

  • Before the 19th century land was a more important factor of production than capital.
  • What matters are differences in the rate of investment, which depends upon the savings rate. Critics of the Solow Model argue the savings rate is endogenous and that people save more when they are rich rather than they are rich because they save more i.e. Solow doesn't explain why there are differences in investment rates.
  • 'Can't afford to save' argument doesn't work for countries just above subsistence. Although perhaps a broken savings line could potentially lead to multiple steady state equilibria.
  • Model doesn't explain long-term growth - as long-term steady state has no growth and shifts in savings rate are only temporary and finite.
  • If the model includes population growth than you can get capital dilution where   Δk=γf(k)-𝛿k-nk where again we can solve for yss=A(kss)^α. However the model, with only a 34% predicted difference, is insufficient to explain the large differences in gdp/capita that we see in the data.