Uses general economic theories (micro-/macro economics), but focuses on a specific goal. Development economics has been available at least since Second World War, maybe earlier depending on how you define it.
Income inequality in the world high today. Historically, countries were more equal. After the industrial revolution, a set of countries took off.
Income differences between countries are often compared with the gini index of the world countries, weighted with population.
To catch up, the poorer countries must develop faster than the rich countries. The poorest nations haven't started doing this until after 2000. Reasons: Economic policies better (better control over budgets, monetary policy, exchange rates, reduced state intervention) and terms of trade became better (prices on local resources grew, such as oil prices).
Income diffrences within countries are also important. There are, however, cases where average incomes rise, but also income inequality within the country.
The financial crisis didn't hit as hard in Africa as here. Per capita income stagnated, but not declined (like in Sweden). Could be because of lack of intergration in the world, weren't involved in the bank. Still suffered from export losses, but not all else.
Geographical location also crucial. Easier to develop if you're close to richer or fast-developing countries (dense agglomerations). Also access to sea etc. Some also theorize that economies focused on natural resources rather than manufacturing don't get as many positive external effects on the rest of the economy.
In 1970, East Asia & Pacific and China had a lower share of global exports than Sub-Saharan (0.022 and 0.006 vs 0.033), but in 2007, they had 0.115 and 0.077 while Sub-Saharan Africa just had 0.018. Modern companies and consumers don't buy goods and services (e.g. outsource) from Sub-Saharan Africa, but they do from East Asia and China. Requires reliability, which depends on a strong infrastructure. North Africa has done better, Marocko, Egypt etc, but very little in Sub-Saharan Africa. Foreign investments that come in usually based on oil and natural resources, which doesn't generate much local businesses oppurtunities. Investments in general different.
Poorest area today includes: Sub-Saharan and South Asia, whose development has stagnated the last 50 years.
On the other hand, we have East Asia & Pacific and China which have been growing like crazy.
Life-expectancy in Sub-Saharan Africa has just increased from 41-51 in the last 50 years, while the fast-developing countries have gone from 39-72.
Many countries are in a poverty trap, that is, a spiral of slow development. Without investments, no growth, without growth, no investments. Must increase K/L. Save and invest enough to replace the depreciation of capital stock, and to get tools in the hands of the new people in the labor market.
Milennium goals, the idea was to give them increased aid, which would start a spiral of fast development. However, today we've seen that even countries with good opportunities, such as a lot of oil, still haven't led to development. Economic environment, policies, corruption etc are also extremely important.
Total factory productivity - Increased output in total. Usually new machines, rationalizations, infrastructure etc(?).
"Natural resource curse" - Sachs and Warner (1995). Sitting on natural resources should be good for you. However, historically, we've seen that natural resource dependent countries have done worse than others.
Washington Consensus - Aimed to handle large budget- and current account deficits, balance exchange rates, stabilize interest rates, reduce state ownership of firms etc. However, difficult because it was directed from Washington but implemented locally. Implementation wasn't succesfull.
Why is productivity
Attempts to go beyond the production function, understand the underlying factors. For example, the Three deep explanations from growth litterature:
Prospects for growth in Africa
Johson, Ostry, Subramian (2007)
Looked at the factors that lead to success in some countries. What was the situation when these countries took off? Institutions, economic environment etc. Made a check-list and compared it to current statuses of African countries. One important condition was a focus on the manufacturing sector.
Optimistic conclusions: the situation is about the same in Africa today as when these countries took off. Some concerns are that the environment is different today, such as higher transactions costs and exchange rates. But still positive look overall.
Y = Output, GDP
K = Capital
L = Labor/Workers
Y = F(K, L)
Constant returns to scale:
F(ZK, ZL) = Z * F(K, L)
Z = 2
F(2K, 2L) = 2 * F(K, L)
Capital per worker (k):
k = K/L
Output per worker (y):
y = Y/L
A = Productivity
F(K, L) = AKα * L1-α
Constant returns to scale:
F(ZK, ZL) = ZF(K, L)
AKα * L1-α = ZF(K, L)
AZKα * ZL1-α = A * Zα * Kα * Z1-α * L1-α
= Z * AKα * L1-α
Because F(K, L) = AKα * L1-α we get:
F(ZK, ZL) = ZF(K, L)
F(K/L, L/L) = 1/L * F(K, L)
F(K, L) = AKα * L1-α
= A(K/L)α * (L/L)1-α
K/L = k = Capital per worker
Y/L = y = Output per worker
y = Akα
Diminishing marginal product
F(K, L) = AKα * L1-α
MPK = (α * F(K, L)) / αK
MPL = (α * F(K, L)) / αL
αL = Small change in L
F(K, L) = AKα * L1-α
MPK = (α * AKα * L1-α ) / αK
MPK = α * AKα-1 * L1-α
AKα * L1-α
α = Share of national income paid to capital store
1-α = Share of national income to ...
A firm = profit maximizer
Maximize profit (π) = Y - wL - rK
Y = Output
wL - rK = Cost
Max π = AKα * L1-α - wL - rK
First order condition:
(1) απ/πK = AαKα-1 * L1-α - r = 0
(2) απ/πL = A(1-α)Kα * L1-α - w = 0
Y = AαKα-1 * L1-α
r = MPK
w = MPL
(A) Total payment to K = r * K = ... [Missing notes]
Created by Robert Solow in 1956
r = investment rate
d = depreciation rate
ΔK = Investment - depreciation
ΔK = ry - dK
K = 100
r = 20%
d = 5%
ΔK = 0.2(50) - 0.05(100) = 10 - 5 = 5
If in year 0 we have:
Kt0 = 100
Next year, we'll have:
Kt1 = Kt0 + ΔK = 100 + 5 = 105
When ΔK > 0, capital stock is increasing
When ΔK < 0, capital stock is decreasing
At ΔK = 0, we have the so called steady state level. Here rf(K) = dK
In the Solow model, all functions are described per worker:
y = f(K) = output
dK = depreciation
ky = Investments
Combined with a Cobb-Douglas production function:
y = A * K^α
ΔK = r * y - dK
Substitute in y:
ΔK = r * (A * K^α) - dK
We know that at steady state ΔK = 0, so we want:
r * (A * K^α) - dK = 0
Move dK to right:
r * (A * K^α) = dK
Divide all by d:
(r * (A * K^α))/d = K
Divide by K^α:
(rA)/d = K/K^α
(rA)/d = K^(1-α)
Put into exponent (1/(1-α)) on both sides:
(rA/d)^(1/(1-α)) = (K^(1-α))^(1/(1-α))
(rA/d)^(1/(1-α)) = K
This is the steady state level of capital (Kss), so:
Kss = (rA/d)^(1/(1-α))
To get the steady state output level (yss) we do:
yss = f(Kss)
yss = AKss^α
Substitute in Kss:
yss = A[(rA/d)^(1/(1-α))]^α
yss = A(rA/d)^(α/(1-α))
yss = A^(1/(1-α)) * (r/d)^(α/(1-α))
We see that with larger investment rate (r), larger productivity (A) or lower depreciation rate (d), both the steady state output level of capital (Kss) and of output (yss) will rise. This can be shown in the Solow diagram by shifts in the various curves.
Let's assume two different countries with same level of productivity (A), depreciation (d) and labor (L) but different investment rate (r) (We're dealing with the long run)
yssi = A^(1/(1-α)) * (ri/d)^(α/(1-α))
yssj = A^(1/(1-α)) * (rj/d)^(α/(1-α))
Note that only investment rate (r) is different.
yssi / yssj = (ri/rj)^(α/(1-α))
α = 1/3
ri = 0.2
rj = 0.05
yssi / yssj = (0.2/0.05)^((1/3)/(1-(1/3))) = 4^(1/2) = 2
yssi = 2 * yssj
A 45-degree line. However, when comparing with actual statistic data, we generally see that the actual difference is below this line:
Three possible reasons:
So we can conclude that the Solow model cannot predict long-run growth rate at steady state. However, it can predict short-run differences in growth.
Δk = rAk^α - dk
Growth rate of k per worker = Δk/k = (rAk^α - dk)/k = k (should be "k hat", i.e. line above k)
k = rAk^(α-1) - d
If country is at steady state (kss), then k = 0
If k > kss, rAk^(α-1) < d, then k < 0, and k is decreasing
If k < kss, rAk^(α-1) > d, then k > 0, and k is increasing
Let's assume two different countries with same A, d and L:
ki = Poor country
kj = Rich country
They'll have the same production function rAk^(α-1), but same d
The poor can grow faster than the rich: ki > kj
Let's now assume different investment rate (r).
They'll have different production functions riAk^(α-1) and rjAk^(α-1), but same d
ki > kj
ri > rj
The model tells us that countries grow differently because they invest differently. But they don't tell us why such differences occur.
There are various explanations:
(1) Saving differences. The same logic can be applied to individuals. Individual income inequalities is based on saving (S) differences.
In a closed economy I = S (the country invests all it saves).
(2) But in an open economy, there can also be:
The Solow states that higher S leads to higher I, which leads to higher Y. Savings is exogenous
However, the cause and effect are not necessarily this obvious. Richer countries might save more because they grow faster, not the other way around. In this case, saving is endogenous because its affected by income.
Can we combine the Solow model with endogenous savings?
Why do poor countries have lower savings rate? Possible reasons:
Differences in saving state:
Let's assume a certain treshold of income and capital at y* and k*
If y < y*, then s1 = r
If y > y*, then s2 = r
s1 < s2
Depending on your income, you save differently.
We have two different functions for the two countries: s1f(k) and s2f(k)
While the poor have a steady state of kss1, the rich will have a higher capital stock at steady state, because of higher savings and thus higher investments.
The poor country is locked into a poverty trap.
Poverty trap: Poor countries save less because they're poor. Small savings make the countries remain poor.
"Big push": Injecting large amounts of foreign aid to poor countries to give them a push into fast development. Problems in reality, maybe not all money go to investments. Or investments might not work in practice, because of lack of property rights, institutions etc.
Human capital and physical capital cannot explain all income differences between countries. How do we measure the remaining part?
Productivity (A) = the effectiveness with which factors of production are converted into output.
Y = Total output
K = Physical capital (e.g. plant & equipment)
L = Workers in the economy
h = Human capital per worker
Y = A * K^α * (L * h)^(1-α)
We need to determine the relative importance of productivity growth A compared to growth in factors of production K^α * (L * h)^(1-α)
Generally we want per capita output, so we divide by L on both sides, resulting in:
Y/L = A(K/L)^α * h^(1-α)
To simplify, we rewrite:
y = Y/L
k = K/L
y = A * k^α * h^(1-α)
This is essentially:
per capita output = productivity * factors of production
To compare across countries, we divide country 1 by country 2:
y1/y2 = A1/A2 * (k1^α * h1^(1-α))/(k2^α * h2^(1-α))
We can move productivity (A) to the left side:
A1/A2 = (y1/y2) / ((k1/k2)^α * (h1/h2)^(1-α))
How do we define alpha (α)? The consensus is that α = 1/3, which is concluded by various data.
max K AK^α - rK
We call Y = AK(?)
α * Y/K = r
α = (r*K)/Y
The last step can be measured by data, which in general shows 1/3.
If α = 1/3, we get:
y = A * k^(1/3) * h^(2/3)
Statistically, we can see that countries defer in factors of production and productivity. Countries vary which is relatively the lowest, which tells us if we should focus on the factors of production or productivity to increase the income in the country the most.
Overall, we see that there are large productivity differences between countries.
Important to be aware of statistical problems ("noise" in the data), which might show up as productivity differences but in reality is just bad data. E.g. human capital as in number of years in schooling can still vary because of quality differences in schooling, which means that differences in factor accumulation across countries would be too ssmall and implied productivity differences would be too large.
Even taking this into consideration, it's still clear that productivity differences are large.
Relative contributions of productivitiy and factors of production per capita. Statistically, we see that the richest country in general share the same level of factors of production per worker, while they defer in productivity. For the poorest countries, we see the opposite, that the factors of production
Let's compare growth differences between productivity and factors of production:
y = A * k^α * h^(1-α)
Rewrite in logarithimic form:
ln y = ln A + α * ln k * (1-α) * ln h
During 1970-2005, the average annual growth rates for the US are:
y = 0.0157 = 1.57%
k = 0.0215 = 2.15%
h = 0.0028 = 0.28%
The reason we take such a long time period is to avoid normal fluctuations, e.g. business cycle.
By ranking countries by growth rate we see that:
Higher growth group: High growth in both production factors & productivity.
Lower growth group: Positive growth in production factors, but negative productivity growth.
If we want to understand why some countries grow faster than others, it seems that we should focus on productivity differences.
Technology = How we use production factors to produce output. A determinant of productivity, just as schooling is a determinant of human capital.
Doesn't come for free, requires research and development (R&D, often denoted by gamma, γ). Generally come from rich countries, poorer countries not good prerequisites for R&D. However, can be spread freely, because it's nonrival. Not like machines, that if I have it you can't have it (rival). Ideas, on the other hand, we can both have at the same time. This is, howoever, also a problem for those that bear the R&D costs, because of the free rider problem, which might lead to less R&D being produced in the world than is demanded.
Two ways how a country can acquire new technology:
Poor countries generally focus on number 2.
Two countries, rich country 1 and poor country 2. Country 1 is the leader (innovator) while country 2 is the follower (imitator). Country 2 can copy the tech from country 1. The leader will have a large share of its population doing R&D, i.e. a cost of innovation (γ).
However, we also have a cost of copying, which generally depend on the difference of technology level between the two countries. Let's assume that the cost of copying is a function of the technology (productivity) gap between the two countries: c(A1/A2). We could call this function μ
Technological growth (A) could be described as A = γ/μ * L.
Follower will have high but declining growth rate in productivity, until they in the long run meet at the same growth rate at steady state (not same level of productivity, but same productivity growth).
Increasing R&D in the leader leads to permanent higher growth rate for both the leader and its followers.
In the real world, not clear which country is the leader. Different countries lead with respect to different technologies. Should not be taken as a literal discussion how the world works.
Barriers to International Technology Transfer:
Zambia has 14% of that in the U.S. Could this be only because of technology differences? Probably not. How many years behind would Zambia have to be for these numbers to make sense? We would get the answer of 294 years. This seems unreasonable considering Zambia still has cars, computers, mobile phones etc.
We want to define efficiency as the effectivenes with which factors of production and technology are combined to produce output. We want to capute anything that accounts for diffferences in productivity other than differences in technology.
A = T * E
A = Productivity
T = Technology
E = Efficiency
To calculate E we must define T as the rate of technological growth and A as rate of productivity.
Comparing US to India, assuming India is 10 years behind in regard to technology level, we would see that the efficiency of India compared to the U.S. is 0.37 (37%?). The technology gap is just 0.94 (6% less than US?).
Conclusion: Unless lags in technology are extremely large, we should focus on efficiency as it seems to be the crucial factor.
Types of inefficiency:
Paper by Hall and Jones entitles "Why do some countries produce so much more output per worker than others?". The conclusion is that the answer can mainly be attributed to differences in productivity, as a result ot institutions and government polities (collectively called social infrastructure).
Main hypothesis: A strong relation between social infrastructure and output per worker.
Per capita output (y = Y/L) can be decomposed into three parts:
After calculating statistics for these components we reach the conclusion that:
Why is this?
Social infrastructure the main reason.
Social infrastructure: Institutions & government policies that procide the incentives for individuals and firms in the economy. E.g. degree of corruption, barriers to trade, contract enforcement.
How do we estimate the effect of social infrastructure? We can define it as an index:
Social infrastructure (S) = (A + B) / 2
A = Index of government antidiversion policies, which would take into account law and order, bureaucratic quality, corruption, risk of expropriation and government repudiation (rejection) of contracts.
B = Index of the degree to which a country is open to international trade.
By doing a regression of this social infrastructure index against per capita output we see that there indeed is a relation. The authors of the paper have focused on instrumental variables (IV) estimates, but we'll look at OLS (ordinary least squares) results which should be sufficient.
The estimated coefficient on social infrastructure (OLS) = 3.29 is highly significant and implies that a 0.1 increase in the index of social infrastructure is associated with an increase in output per worker of about 0.329 log points, about exp(0.329)-1 = 39%.
Conclusion: Differences in social infrastructure across countries cause very large differences in per capita income across countries.
How do we know that there isn't a reverse causual, i.e. that rich countries simply have much better institutions, not that better institutions leads to riches? We use instrumental variables (IV), i.e. extra variables that are correlated with social infrastructure but uncorrelated with the residual E in the per capita output equation.
Choice of instruments:
It turns out that the results are fairly similar, which would not be the case if rich countries would be the cause of better institutions.
GDP != GNP
Markets more integrated:
Effects of openness on growth
Definition of openness:
There's a general agreement that openness and growth are positively correlated. But is openness a cause of growth?
The conclusion is that openness is good for growth.
Effects of openness on productivity
Openness gives access to new technology, knowledge etc. One channel is trade. Imports give you access to machines with new technology built in. But not only imports leads to productivity increases, but also exports. Information exchange in general.
Still cause and effect problem.
Competition as a result of openness is also good for productivity, because companies get an incentive to become more productive. Even if it doesn't lead to productivity increases in domestic companies, bad companies will be knocked out while good and efficient stay. Good for consumers.
Effects of openness on factor accumulation
Mainly foreign direct investments: Foreign companies invest in the country.
But also loans and foreign aid.
Growth with capital mobility in fully open vs closed economies. Small open economies, ie price of capital is given (the small country cannot influence market prices).
Production function for open economy:
y = Output per capita (Y/L)
A = Productivity
k = Capital per capita (K/L)
y = Ak^α
0 < α < 1
Marginal product of capital (MPK) = αAk^(α-1)
Capital price (r) = MPK
Because α < 1, if k goes up, the more MPK decreases.
Because we're dealing with a small open economy, the prices in the country = world market prices:
Capital price (r) = World capital price (rw)
So we get:
rw = αAk^(α-1)
k^(α-1) = rw / (αA)
k = (rw / (αA))^(1/(α-1))
k = ((αA) / rw)^(1/(1-α))
The capital per capita ratio will depend on world market price (rw). If price of capital goes up in the world, less capital will be invested per capital in Kenya.
Isn't influenced by savings.
We can plug in our k expression in y:
y = Ak^α
y = A(((αA) / rw)^(1/(1-α)))^α
y = A^((1-α)/(1-α)) * A^(α/rw) * (α/rw)^(α/(1-α))
y = A^(1/(1-α)) * (α/rw)
We still see that the main thing that matters for output is rw.
However, we must separate GDP from GNP.
In this case y = GDP
If Savings (S) > Investments (I) we get that GNP > GDP
In Sweden, we've been saving more than we've been investing, trying to pay off dept. Improved our international asset position.
But for a closed economy:
Savings also influence k and y. Savings = Investments
They looked at global data on savings and investments to determine degree of openness, and found that investments are highly influenced by savings.
Correlation between savings and investments:
0 = No correlation
1 = Full correlation. Savings = Investments
In 1964-1980: 0.89
In 1990-1997: 0.60
The correlation has declined, because economies are more open, but there still is a large correlation.
There's also information advantages of domestic investments, so savings might never be equal to investments, but still might decline.
On the whole, developmental economists believe that free trade is good. But financial integration is more controversial. Theoretically should be good to maximize world GDP, but it's more complex, because it also exposes you to international shocks, and you also lose some macro economic tools, such as monetary policies etc.
Opposition to openness
Free trade and capital mobility is good in aggregate terms, i.e. for the whole of the world. But maybe not for each individual sector, area or even country.
In the real world, we have frictions and switching costs, such as when stopping an industry we have a lot of unused machines.
However, if we distort the market we'll have a smaller pie to share.
Paper: "Aid effectiveness - opening the black box"
Is aid good for growth? Same problems as with openness.
We're talking about aid as ODA (as defined by DAC)
Uses cross-country regression framework to find out if there's a causual effect between aid and growth. Hard to get consensual results.
If you try to redefine aid as the growth-significant part only, then you might be able to isolate its effect.
Seldom people consider aid crucial for growth. The results usually give half a percentage difference in growth or something like that.
Aid is considered a black box. We know nothing of what's going on, but put in money and hope for the best.
The paper tries to go past this by defining the detailed mechanisms of aid:
North Korea vs South Korea. Had the same structure and growth until South Korea went for democracy and an open economy.
Many small firms that produce very similar goods
Perfect competition leads to P = MC, i.e. the resources are allocated optimally.
Production possibilities frontier - Combinations of goods and services that can be produced if all resources are used efficiently.
When are resources not used efficiently?
Institutions: The rules of the "game" in a society. The humanly devised constraints that shape human interaction. Consists of:
Organizations: The "players", or the humans that interact in society.
Good institutions favor socially beneficial activities (productivity/competition) rather than rent-seeking (seek redundant profit?).
Total production (Y) is determined by:
All of these can be affected by governmental institutions:
Factors of production
Rule of Law:
Statistically, we can see that rule of law is related to income per capita. Countries with high income generally have better rule of law. Also same relation with factors of production and productivity.
With better rule of law:
Why do bad institutions hurt economic growth?
State enterprises: Political control of key firms (banking, heavy industries) which are then run inefficiently.
Marketing boards: Through which farmers had to sell their products. Opportunities for corruption, and didn't give farmers the higher prices they expected.
Trade polities: Import substitution, export
Even if you have good ideas, it's hard to realize them if the institutions are bad.
Tasks of institutions: Create, regulate, stabilize, legitimize markets.
Several types of instutitions are needed for long-run economic development.
One size does not fit all. Might not be possible to create "blueprints" that works for all countries. Policies that improve on type of institution might also destroy others which as a whole might not be growth-enhancing, e.g. reform programs like the SAPs (structural adjustment programs).
How do we explain the positive correlation between income and institutions?
Causation? Do good institutions lead to higher income or income lead to good institutions? Probably both. The question is which come first, or which should be prioritized.
History also important, 22 of the 30 most corrupt countries in the world are former European colonies:
Colanization -> Bad governments -> Low income
Bad institutions are not always an impediment to growth (Europe/US started out with pretty corrupt governments).
Not the main determinant of income, but still
Income per capita vs income of the bottom quintile shows that there's a clear relation between average income in the whole country and the average income for the poorest group. There are, of course, different types of growth that have different effects on income for the poorest and differences between countries, but in general, to improve poverty you must improve the average income level in the country.
The same thing goes for income redistribution, that it could be especially important for some countries, but for most country poverty reduction is mainly needs growth.
Some studies have showed that inequality is generally consistent during growth, which means that the poorest groups will benefit from increased overall growth.
Sources of inequality
Statistics have shown that the highest return come from tertiary education, rather than secondary or primary, although the latter are probably required for tertiary and good for a number of other things.
There are two statistically observed effects of education, that could be important for deciding policies for inequality reduction:
Returns to education is calculated by looking at different levels of education between people and the income they recieve afterwards in combination with the lost years of income during education.
Factor price equalization: The relative prices for two identical factors of production in the same market will eventually equal each other because of competition. That is, cheap labor in poor countries is more demanded than expensive labor in rich countries, which will lead to jobs moving to poor countries until this equalizes.
Marginal products of physical and human capital: Two ways of investing resources: Physical capital (buys machines) or human capital (educates himself). Physical capital will give an even return (marginal product) while human capital have deminishing returns. In poor countries, the return of human capital is higher than physical capital. Growth today is more skill-intensive than it used to be. An explanation why inequality might be less important today in fighting poverty. Also, if investments are not determined by domestic savings but international savings, it might be even less important.
Is taxing the rich a viable solution?
Relationship between income inequality and the desired tax rate: Assumes proportional tax. The persons with median income decide the tax rates. The lower the median income, the higher the tax rate. If inequality goes up, the taxation pressure will increase and the median income will fall. High taxes reduce investments, which might slow growth.
Inequality might sometimes be good for incentives, but inequality could also lead to increased social tensions.
Mobility also important:
Society should be given a proper education so that they can move up in social groups.
Intergenerational income mobility in Canada: Large mobility between income groups, because the society allows it.
Paper "Inequality among world citizen" (Bourguignon and Morrison):
Tries to calculate gini for world popultion between 1820-1992. The result shows that inequality went up until about 1950 and then stagnated. Maddison computed world GDP over this period, then used income levels (per decile, or actually for 11 income groups, separate for 90-95% and 95-100%) for as many countries they could find. Six regional blocks. Estimates:
Gini, Moderate poverty, Extreme poverty
1820: 0.5, 94.4, 83.9
1950: 0.64, 71.9, 54.8
1992: 0.657, 51.3, 23.7
A period of drastic reduction of global poverty.
Using the Theil index, compared inequality within countries, between countries and total:
In 1820, Inequality was generally a result of domestic income inequality rather than inequality between countries.
But in 1950-1992, the gaps between countries is now the main explanation of world inequality, while domestic income inequality declined.
Can culture explain why some countries are rich while others are poor?
Culture: The values, attitudes and beliefs prevelant in a society.
Culture, when defined this way, apparently affects how people act. We want to identify regularities, where the majority of people share some kind of values or beliefs.
Culture can be seen as a set of social norms or informal institutions
"Protestant ethic": Max Weber claimed that a key cause for why northern European economies grew rich from the 16th century onwards was due to the "protestant ethic" (i.e. celebrating hard work and accumulation of wealth)
"Asian values": A fondness of strong and stable leaders, social harmony, collectivism and economic rather than political well-being. Sometimes argued as a reason for the success of the Asian Tiger economies compared to African economies.
However, difficult to quantify and measure, thus hard to prove right or wrong. Without quantification you cannot test different arguments and see which has the strongest support in reality.
Weil defines six cultural traits important for growth:
Can geography explain why some countries are rich while others are poor?
Geography is fixed and not something a country can change. Instead you have to adapt to your environment and make the best of the situation.
Distance from equator seems to strongly affect income, both towards to the south and north, which suggests that there is a geographical component. Poorer countries are more often in the tropics, while the temperate zones are richer. Even within Africa, countries get poorer as they get closer to the equator.
Climate affects economies in various ways:
Access to the ocean: Necessary for countries to be a part of the global economy. Leads to bigger markets, specialization, exploiting comparative advantage and diffusion of knowledge.
Role of neihbours: Can have both positive and negative effects
"Landlocked" countries are countries with no access to the Ocean. In Europe, landlocked countries are still okay because of good neighbours. But landlocked countries with bad neighbours have had a very hard time to grow.
Guns, Germs and Steel: Diamond (1997) claims that geographical facts have determined the fates of societies. Historically, 10500 years ago in the fertile crescent and the Neolithic revolution (humans transformed from nomadic hunter/gatherers into sedentary farmers). This lead to more dense populations allowing more social stratification (specialization, organization etc). Eventually, this lead to capitalism and the industrial revolution. This is an explanation why this happened in Eurasia and not America. Knowledge could spread more easlily between countries in Eurasia, and in the end America & Africa were colonized. But still doesn't explain why Europe and not Asia became rich in the end.
Another explanation for Europe's success is that Europe has a fragmented geography, and fierce competition between many small states. Many relatively small competing states separated by natural barriers, 500 political units by 1600, and 25 states by 1900. Some consider China having four bigger "natural" units, while it was completely united in 221 BC, which might have meant less competition.
Geography fragmentation can be both:
Bad: A unified state implies a bigger market, more specialization, easier spread of ideas and less war
Good: Competition between states, acting as a disciplining device. Can discourage corruption and dictatorship. Small and more efficient governments.
A drastic example: China banned ocean-going ships in 15th century, but when Columbus was rejected by many he could find support form the Spanish. If Europe had been like China, Columbus may bever had set sail.
Diamnod's story (1997): Some societies developed earlier and have since had an advantage.
Acemoglu et al. (2001) story. Colonizes set up different institutions in areas with different geographical characteristics.
Sachs' (2001) story: Some areas have always been poorer, and have therefore had worse institutions.
Obviously, one story doesn't exclude another, and there are many factors determining growth, with causation often going in both ways. The question is which is the most important. Where should money be spent, to give the highest possible return?
Rodrik and Subramanian (2003) combine geography, trade, and institutions in one regression:
Institutions = Rule of Law
Integration (trade) = X-M / GDP
Geography = Distance from the equator
ln(GDP/capita) = alpha + beta0 * Institutions + beta1 * Integration + beta2 * Geography + E
Once the positive effect of institutions is accounted for neither better geography nor more trade is associated with higher incomes.
Institutions seems to be the most important.
Natural capital: The value of a country's
Productive, produced, limited used, earns a return, wears out.
The difference is that natural capital is not produced. Instead, it's used to produce physical capital, with the help of human capital.
GDP (Y) = Sum of goods/outputs. Output is produced by combining human capital (labor, K) with physical capital (K).
Y = f(L, K)
Land: A lot of workers, farmers that produce for themselves.
Resources: Concentrated ownership, few workers.
Market - Forward linkage: Local supply of other goods. Production for another company(?) Products used locally.
Market - Backward linkage: Local demand of output. If products are bought abroad, there's no local demand.
Political: Increases in government budget, tax revenues.
In cross-country comparison there's a negative relationship between GDP and natural resources. How come?
Probably more channels:
Market - Dutch disease: When exporting resources, you appreciate the exchange rate. This makes it more difficult for the manufacturing sector (often big in developing country) to export, which causes losses to GDP.
Political - Government behaves myopic: Over-consumption of budget.
Political - Corruption: The increase income doesn't lead to more income for the population but disappears in corruption.
Politlcal - Conflicts: For example, rebels control certain areas of the country, financing themselves by controlling natural resources. Or rebels attack oil fields.
Mehlum, Moene and Torvik (2006) states that the negative correlation between natural resources and GDP depends on institutional quality. No relation if there are good/strong institutions, but negative if there are bad/weak institutions.
Unclear policy implications: Should countries avoid finding natural resources??
Example of within-country comparisons studies:
Nigeria. What is the impact of natural resources (oil)?
How would GDP evolved in absence of its oil endowments?
Counter-factual: Is there a situation/country similar to Nigera but without oil?
Cross-country comparison: A neighbour. Might still be too different for direct comparisons.
Within-country comparison: Nigera before oil discovery(?), or different regions within Nigera. Hard to control for Dutch disease, civil wars or central government behavior (as they affect the whole country)
Study 2 - USA:
Backward linkages, Black et al (2005)
Price of oil goes up -> demand of coal increases (substitute goods) -> price of coal goes up -> demand of coal workers goes up -> earnings in mining sector goes up.
Impact on non-mining sector?
Compare coal-rich to coal-poor sectors (counties)
Does econometric comparison on the data. We see that the employment in non-mining sector also increases as coal prices increases. Difference between traded/export-driven and local/non-tradeable sector (construction, services, retail). Traded sector not so affected, but local sector strongly so. The bust is worse than the boom.
Tradable sector goes down if workers move to non-tradable or exchange rate appreciates
Study 3 - Peru:
Backward linkages, Aragones and Rud (2009)
Compare households that live close to the mine and those living far away.
If close to the mine:
We find evidence for backward linkages in many countries.
Study 4 - Brazil:
Political - Public expenditure mechanism, Caselli and Michaels (2009)
Does the finding of oil have a positive effect on household welfare? Or is there a natural resource curse?
Is there evidence of increases in public budget and expenditure as a result of finding oil?
Oil production goes up -> Tax revenues goes up (oil is taxed) -> Government expenditures goes up -> What happens to the households welfare?
We must exclude all other channels to focus only on the political.
That's why the pick offshore oil fields, so that it doesn't affect the local market. The only way the oil fields interacts with the households is through the government.
Positive impact on gov revenues
Positive impact on gov expenditure
Disappointing impact on HHs welfare measures
There's a miscorrespondance between twhat the government report it's doing, and what is actually happening.
Study 5 - Brazil:
Political - Public expenditure + election mechanism, Moneiro and Ferraz (2009)
Is there evidence of misreporting/corruption in areas with oil revenues?
Do politicians use natural resource revenues to buy wealth or their own re-election?
Buying re-election: Visible changes just before election. Patronage, give benefits to friends or supporters. Vote-buying: Give someone money to get a vote (that's why it shouldn't be public what someone votes for, so it cannot be controlled).
Are politicians good or bad?
Results - Are politicians "bad"?:
Study 6 - Colombia:
Political - conflict, Angrist and Kugler (2008)
Is there evidence of increases in conflict when finding natural resources?
Colombia major exporter of coca to the US.
What decides where it's good to produce coca? Climate, soil, politics or infrastructure
Compare coca producing areas with non-coca producing areas.
Results for coca producing areas:
Net impact of increase in coca production can be considered negative.
Study 7 - Colombia:
Political - Conflict, Dube and Vargas (2008)
Is there evidence of increases in conflict because of coffee or oil production?
Why do people become rebels? It's like a job, they get food, living, sometimes even wages. Also possibility to threaten people, earn money. Or even belief in political power. Big organizations.
A possibile relationships:
Price of coffee goes up -> more people become farmers
Price of coffee goes down -> more people become rebels
Different areas that produce coffee and that have high oil production (measured by closness to oil pipes).
Results for coffee producing municipalties(??):
Results for oil producings:
Putting the two things together:
Positive impact of oil production on paramilitary attacks
Negative impact of coffee production on all outcomes
Main reference: Michaels (2010) "Challenges for research in resource-rich countries"
[TODO: Check missed excercise lecture]
[TODO: Check missed lecture notes]
Dependancy ratio - People of working age relative to people of non-working age. A ratio of 1 would mean one worker per non-worker, which is very high. Index calculation: Below 15 and over 64 divided by those aged 15-64.
Statistics on HIV available from: US Aid - Demographic and health survey
HIV positive women generally have more children in younger years but later in negative years than those that are HIV negative.
There might be behavioural effects on general population.
Different opinions in litterature:
Young (Year?) argues that HIV causes you to have less kids because you want to avoid infecting people/your children with HIV.
Kalemli-Ozcan (2006, 2010), Lorentzen et al. (2008): Young is wrong, fertility increases because of adult mortality
Both mechanisms might actually be true.
Durevall and Lindskog (2006) in Malawi: HIV reduces fertility slightly due to both biological and behavioral reasons (The strong negative impact on fertility that Young reports has not been reconstructed, but a slight effect is visible). The effect is small because older women have less children, younger women have more children (or give birth earlier). Dependency ratio does not improve.
Reduced life expectancy reduces expected income. Many kids repeat or drop out from school. Why go to school if I will die early? (Many people even overestimate their likelyhood of dying early).
Education that informs of sexual diseases might be important for prevention, but are lacking in many areas in practice. Sex is somewhat tabu in many areas, and also difficult to influence in a society where you generally marry and have kids young, which makes it harder to advocate condom usage or sexual abstinence in younger years. Also illness and death among teachers (and incentive to not report it), and bad school management.
Fortson (2006): Less average years of schooling in countries with higher rate of HIV
Beegle et al (2006): Child in Tanzania who loses mother early in life has one year less of schooling and is 2 cm shorter.
Income distribution and poverty
HIV/AIDS might raise poverty even if impact on GDP is small (Salinas and Hacker, 2006)
Many households that become poor will become chronic poor.
UN millennium goals of reduced poverty are probably in jeopardy in several countries due to AIDS.
Strong income inequality, a few who are very rich.
Food price inflation might be more important.
To analyze effects on income and distribution you must look at the household composition and follow them for a number of years. The solution has been to support households that takes the hardest hit, when both parents have died etc.
"Tobin tax" and back to Bretton Woods?
Politicians should regulate bankers rather than the other ways around.
Find a balance between markets and regulations (which are both imperfect).
Industrialization not an option for developing countries, because the RoW is already so far ahead in efficiency. Must instead develop through natural assets. However, this requires functional institutes regulating property rights.
Propinquity = We care for the people close to us. Leads to less sharing with the rest of the world. Also leads to overconsumption today, since we don't know the people of the future.
Natural assets curse?
Negative effect of natural assets:
Sierra Leaon - A lot of diamonds, but mainly lead to civil war
Positive effects of natural assets:
Botswana & Norway
Difference between agricultural commodoties and non-agricultural commodities (e.g. oil) because the first are renewable.
What differentiates the prospect of Middle-East compared to Africa?
Good governance, or the harnessing of natural capital in productive sectors like infrastructure.
"Natural assets are living dangerously: lacking natural owners they are liable to be plundered"
The value of natural assets should be captured by government, not third-parties.