Chapter 2' Trade and distribution

1
Chapter 2. Trade and distribution
Course FB, Winter 09
2
Outline

• Some basic evidence
• The debate on the source of increasing inequality
  and mounting unemployment in developed countries
  trade, technical progress, or demographics?
• Naive applications of the 2x2x2 Hecksher-Ohlin
  model with skilled and unskilled labor
• A 'sensible' assessment of the causes of
  increasing inequality by applied general
  equilibrium models of the world economy
• Limits to redistribution and gains from trade
• Poverty and trade the key relationship between
  trade and growth

3
1. Some basic evidence apparent effect of trade
and globalization variables on within country
inequality

• Evolution of earnings inequality
• Evolution of GDP labor share
• Skilled vs. Unskilled debate
• Simple regressions relating inequality and trade
  indicators

4
(No Transcript)
5
Source, IMF, WEO, 07
6
Source, IMF, WEO, 07
7
Source IMF, WEO, 2007 18 advanced countries
1982-02
8
Source, IMF, WEO, 07
9
51 countries, 1981-03
Source, IMF, WEO, 07b
10
51 countries, 1981-03
Source, IMF, WEO, 07b
11
Source, IMF, WEO, 07b
12
2. The debate on trade related inequality and
unemployment

• Absolute and relative fall in bottom real wages
  during the 1980s in the US debate on the
  possible effects of North-South trade on wages
  and employment in the mid 1990s (Freeman, 1995)
• Debate older than this e.g. employment effect of
  trade with developing countries Martin and Evans
  (1981), Balassa (1986) still very prominent
  today with debate on the effects of "offshoring"
  and "outsourcing"
• Analytical divide occurred in the early 1990s
  about the effect of North-South trade on labor
  demand and skill differentials in OECD countries.

13
Divergent views on the effects of North-South
trade upon labor demand employment and skill
differential

• Labor economists' approach
• Effect of increased LDC trade on skill
  differential is negligible (Berman, Bound and
  Griliches, 1994) versus 20 of actual increase
  (Borjas, Freeman and Katz, 1992).
• Trade economists' approach
• Changes in relative prices too limited to
  generate any significant effect of North-South
  trade on the economy (Lawrence and Slaughter,
  1993) versus impact of trade on skill
  differential is substantial (16 of the observed
  increase or more) Leamer (1994), Wood (1994)

14
Towards a synthesis?

• Synthetic view (Krugman, 1995 Cline, 1997)
• Increasing North-South trade is one of the
  factors explaining changes in labor demand and
  skill differentials in OECD countries but not the
  most important one (less than 20 per cent of
  observed phenomena)
• Other factors include technical progress,
  migrations, labor supply shocks, ...
• Recent contributions would confirm this
  conclusion.
• Very much affected by the fact that skill
  differentials tended to level off - except at the
  very top of the distribution.
• Focus increasingly on the effects of technical
  progress

15
2. A simple Hecksher-Ohlin model with skilled and
unskilled labor implications and limitations

• Basic assumptions two goods, two factors
  (skilled and unskilled labor), two countries,
  perfect domestic mobility of factors, perfect
  international immobility of factors, identical
  homogeneous productions functions, perfect
  competition on domestic markets for goods (see
  standard textbook on international trade, e. g.
  Bhagwati et al.).
• Notations
• Production functions sector i 1, 2 country j
  a, b
• 
  
  
  
• where Q output, L total employment, S
  skilled employment, total labor force,
  skilled labor force, and lower case symbols are
  scaled by the total labor force or employment.

(1)
16
Production equilibria

• Profit maximization
• 
  
  
• where wu is the wage rate of unskilled workers
  and ws the skill premium
• Labor market equilibria
• After resolution

(2)
(3)
(4)
17
Autartik equilibria

• (Aggregate) demand (representative agent)
• where is a vector of preference
  parameters
• General equilibrium
• After resolution

(5)
(GE)
(6)
18
Free trade equilibrium

• Domestic prices within countries a and b are the
  international prices
• New general equilibrium condition writes at the
  world level

(7)
(FT)
19
Some properties of the free trade equilibrium

• Relative factor intensity
• From (2) it comes that the skill intensity of the
  production of good i, si, is given by the
  equation
• Good 1 is said to be relatively more 'skill
  intensive' than good 2 if s1 gt s2 for all values
  of the skill differential, ?.
• Hescksher-Ohlin theorem. At free trade
  equilibrium each country exports the good
  relatively more intensive in the type of labor
  relatively the most abundant in the country.

20
Some properties of the free trade equilibrium

• Stolper-Samuelson Theorem
• Protection (free trade) in a country reduces
  (increases) the return to the relatively most
  abundant factor and increases (reduces) that of
  the relatively scarce factor.
• Protection reduces the production of good 1 (less
  exports) and increases that of good 2 (less
  imports). The relative demand for skilled labor
  declines and therefore its relative remuneration.

21
Some properties of the free trade equilibrium

• Rybczynski Theorem (Closed or world free-trade
  economy)
• "An increase in the quantity of one factor will
  always lead to a worsening in the terms of trade,
  or the relative price of the commodity using
  relatively much of that factor"
• Here
• Increase in the quantity of skilled labor reduces
  the relative cost of that factor and the relative
  price of the commodity relatively intensive in
  that factor.

22
Some properties of the free trade equilibrium

• Factor price equalization theorem (Samuelson)
• Under HO assumptions, in absence of complete
  specialization of trade partners and in absence
  of factor intensity reversal, factor prices are
  identical in the two economies at the free trade
  equilibrium.
• Proof. From (2)
• It follows that and are identical in
  countries a and b and depend only on
  international prices. This implies

23
Controversy about factor price equalization
theorem

• Often presented as paradoxical and unrealistic
  prediction shedding doubt on the whole HO
  framework.
• Some generalization is possible to the case of
  more than 2 commodities and 2 factors.
• Key assumptions clearly are identical production
  functions, absence of commodity-specific factors,
  incomplete specialization, perfectly competitive
  factor markets. But these assumptions are also
  key for the whole HO framework.
• At the same time, the FPE theorem regained very
  much audience lately, precisely because of the
  simultaneity between the growth of North-South
  trade and the increase in the skill diferential
  in developed countries.

24
Explaining the increase in the skill differential
by pure trade theory

• Direct application of the Stolper-Samuleson and
  Rybczyinski theorems within the skilled-unskilled
  labor HO framework
• Less protection in OECD countries (various recent
  trade negotiation rounds) affects positively
  skilled labor, which is the relatively abundant
  factor
• Incorporation of new countries in international
  trade e. g. China - equivalent to relative
  increase in the world supply of unskilled workers
  with a negative effect on the relative price of
  unskilled labor commodities and on the relative
  reward to unskilled labor worldwide.
• First steps towards Factor Price Equalization?

25
Objections to and confirmations of the trade
explanation of the increase in skill differentials

• HO framework changes in the structure of
  activity and labor demand should be the main
  consequence of increased trade.
• Berman, Bound and Griliches (1994) no big change
  occurred in the US.
• Borjas, Freeman and Katz (1992) increased trade
  had modified the 'effective' relative demand for
  the various types of labor (leading to an
  increase in the skill differential equal to 15-20
  percent of what was observed).
• HO framework increased North-South trade should
  reduce the relative price of unskilled labor
  commodities and the skill intensity of the
  production of all commodities
• No big change in terms of trade in OECD countries
  and skill intensity increased rather than
  decreased (Lawrence and Slaughter, 1993)
• Leamer (1992) different taxonomy of commodities
  Wood (1994) some goods imported by Northern
  countries are not produced anymore there.
  (Leamer trade-caused increase in the skill
  differential equal to 20 per cent what was
  observed Wood almost 100 per cent!)

26
Objections to and confirmations of the trade
explanation of the increase in skill differentials

• Why did the skill differential initially increase
  more in the US and UK than in continental Europe?
• Flexible vs. fixed wages. Unskilled unemployment
  rose in continental Europe instead of unskilled
  wages falling. (Krugman, ??). But increase in
  North-North trade, initial effect of different
  labor market closures on the structure of
  activity, different sensitivity to Southern
  trade. (Atkinson, 2000)
• Why did the skill differential increase in
  Southern countries, in contradiction with the
  simple HO framework?
• Definition of skilled and unskilled workers and
  functioning of labor market. Higher unskilled
  wages in formal than informal labor market. Skill
  differential to be measured within the formal
  sector only.
• How should trade deficit (surplus) be taken into
  account? Non-traded sectors? Factor market
  imperfections? Factor supply responses? ...
• Is it always possible to disentangle trade and
  globalization from other forces pushing unskilled
  wages (and employment) down? Is technical
  progress or migration independent of foreign
  trade?

27
3. A "sensible" assessment of the effects of
trade on inequality through applied general
equilibrium models

• Clearly the 2x2x2 HO framework with skilled and
  unskilled workers is inappropriate in view of a
  much more complex reality.
• Hence the idea of building numerical simulation
  models to evaluate the potential effect of
  increased North-South trade on inequality (i.e.
  Skill differential).
• Several world-wide applied general equilibrium
  models available with varying degrees of
  complexity. Some of them are actually close to
  simple generalizations of HO framework.
• What follows is drawn from simulations with the
  model TIDE (Trade and Income Distribution
  Equilibrium Model (Cline, 1997) that generalizes
  Krugman (1995).

28
Specification of the model

• 13 'countries' (groups of countries) US, Canada,
  EU, Japan, rest of OECD, Mexico, rest of Latin
  America, China, Asian tigers ( HK, Singapore,
  South Korea Taian), India, rest of Asia, Eastern
  Europe and rest of the world.
• 3 factors unskilled labor, skilled labor,
  capital
• 5 sectors in each country 3 tradables (skilled
  labor commodity, unskilled labor commodity,
  capital intensive commodity), non-tradable
  intensive in unskilled labor, non-tradable
  intensive in skilled labor
• In each sector, production functions are CES.
  They differ across countries only by a specific
  scale factor that changes exponentially over
  time.
• Preferences are Cobb-Douglas in each country
  (perfect aggregation property) with budget shares
  that are country specific
• Trade is balanced (export of country i in good j
  sum of imports of other countries in good j).
  However, 'effective' import for consumption can
  differ from 'gross' imports because of
  transportation cost and trade barriers (no formal
  modeling of tariffs)

29
Specification of the model

• Factor supplies are exogenous in all countries
• 'Calibration' based on 'plausible' budget and
  factor shares and arbitrary values (elasticity of
  substitution of CES production functions .7 or
  1)
• Transportation costs taken to be proportional to
  distance, tariffs are close to mean tariffs in
  the various groups of countries.
• Country-specific scale factors in production
  function calibrated so as to generate a trade
  matrix reasonably close to the one actually
  observed.
• Calibration undertaken for three different years
  1973, 1984 and 1993

30
Results of simulations deviation of factor
prices from baseline for selected experiments and
countries (percent) (Cline, 1997)
31
Concluding comment on the trade explanation of
the increase in the skill differential

• How much of the observed increase in skill
  differential (and unskilled labor unemployment)
  in OECD countries during the 1980s/1990s was
  caused by the increase in North-South trade?
• Ambiguous and/or heterogeneous econometric
  evidence on relationship between skill
  differential and trade-related variables
  (structure of economic activity, terms of trade)
  identification often requires a restrictive
  theoretical framework.
• Simulations based on less restrictive synthetic
  representation of global economy suggests that
  trade liberalization and expansion explains part
  of the increase in skill differential, but
  unexplained residual is very large.

32
Concluding comment on the trade explanation of
the increase in the skill differential

• In summary
• Yes, the expansion of trade played a role in
  explaining changes in skill differential
• But trade is only one of numerous factors that
  affected that evolution
• Other factors include skill-biased technical
  change, migration and changes in factor supply
  (exogenous or endogenous?), changes in
  functioning of labor-market (de-unionization),
  etc..
• Among them, skill-biased technical progress seems
  the most important one, although its impact is
  uneasy to measure (Krueger, 1993 Acemoglu, 2002)
• Possibly sizable residual

33
4. Limits to redistribution and gains from trade

• The 'gains from trade' efficiency argument
  applies to a representative agent
• Trade permits consumption of the representative
  agent to be above the production frontier
  (argument generalizes to trade liberalization)

x2
P
C
C
x1
p1/p2
34

• Argument generalizes to trade liberalization (T
  tariff)

x2
p1(1T1)/p2
p1/p2
P
x1
35
Case of heterogeneous agents and no available
lump-sum redistribution instruments

• But what if domestic agents are heterogenous in
  terms of factor supply and/or preferences and do
  not benefit all from trade?
• Example unskilled workers in OECD countries
  when North-South trade is liberalized
• If lump-sum redistribution is possible then gains
  from trade at the aggregate level implies
  Pareto-superiority of liberalization
• Things are different if lump-sum redistribution
  is not available
• Redistribution now entails an efficiency loss
  that reduces and, in some circumstances, may
  possibly offset the aggregate gains from trade
• The Pareto-superiority of trade liberalization
  may ultimately rely on the nature of the
  redistribtion investments available.

36
Example of efficiency loss due to redistributing
the gains from trade (inspired by Spector, 2001)

• Standard HO framework with skilled and unskilled
  labor but with endogenous factor supply
• International prices of the skilled and unskilled
  intensive goods ps, pu. Domestic economy
  relatively rich in skilled labor.
• Proportions of skilled and unskilled workers ps
  pu 1
• Labor supply of skilled and unskilled workers
  ls, lu in 0,1
• Preferences of workers i ( i skilled or
  unskilled)
• and after maximization with respect to
  consumption
• where yi is after tax income.

37
Equilibrium without taxes

• Max W(yi, p, li)/ yi wi.li ? li(wi, p)
• Factor supply in the domestic economy ps ls(ws,
  p) and pu lu(wu, p)
• Open economy production equilibrium ? factor
  prices determined by international prices ws
  fs(ps/pu), wu fu(ps/pu)
• International equilibrium solved by equating
  supply and demand of one good as in standard
  model with an additional source of flexibility
  coming from the endogeneity of labor suply.
• Let ps/ pu be the world equilibrium price
• Now, consider some liberalization in the world
  economy that leads to an increase in ps/ pu and
  therefore an increase in ws and a drop in wu.
  There is a positive aggregate gain for this
  economy, since it is relatively rich in skilled
  labor.
• Distributional issue skilled workers are better
  off, but unskilled workers are worse off. What
  should be the redistribution policy?

38
The challenge of redistribution policy

• Suppose that lump-sum redistribution cannot be
  enforced for instance because skilled and
  unskilled workers can only be distinguished on
  the basis of their income.
• Assume the government can apply a non-linear
  income tax to redistribute across the two groups
  of workers.
• The objective of the government is then to
  redistribute from skilled to unskilled workers so
  as to make everybody better off after the change
  in international prices.
• The constraint is that the tax schedule must be
  'incentive compatible' skilled workers must not
  find it in their interest to 'disguise' as
  unskilled by reducing their labor supply so as
  to have the same level of income as unskilled
  workers.

39
Some characteristics and properties of the
redistribution policy

• The government must choose a non-linear income
  tax schedule, that includes a net transfer, t,
  paid by skilled workers and redistributed to
  unskilled workers.
• Through the shape of the non-linear income tax
  (marginal tax rates, tu, ts), the government can
  actually control the labor supply of the two
  groups see optimal taxation literature
  (Mirrlees, 1973)
• The constraints of the redistributive taxation
  policy by the governement can thus be written as
  follows.

40
Explanation of constraints and basic issue

• Welfare of unskilled workers after redistribution
  must be at least equal to initial situation.
• It must not be in the interest of skilled workers
  to appear as if they were unskilled i.e.
  getting the same pre-tax income as unskilled
  workers
• Issue not sure that there exist t u and t s in
  0,1 and t gt 0 such that these constraints are
  satisfied!

41
Efficiency loss

• Preceding framework not complete in the sense
  that there is no optimizing behavior by the
  government.
• There are three degrees of freedom, lu, ls (or
  corresponding marginal tax rates) and t, and two
  constraints. Fully identifying a redistribution
  policy requires setting some kind of objective by
  the government.
• Maximizing social welfare would reduce to
  maximizing Ss (since is fixed). But then,
  the issue arises of what was the optimal
  redistribution in the initial situation (value of
  ).
• Another objective would be to minimize the
  efficiency loss, as measured, for instance, by
  total output, and to measure the extent to which
  the fall in output due to redistribution offsets
  the gain arising from the gains in terms of
  trade.
• (Not done here because requires closed form
  resolution of the whole model including taxation
  numerical illustration suggested).

42
Optimal redistribution

• Spector (2001) uses this framework to analyze the
  gains from trade under the assumption of a
  government maximizing social welfare in autarky
  and in free trade equilibria.
• His conclusion is that, for a small economy, free
  trade may lead to a drop in social welfare if
  equilibrium prices at the free trade equilibrium
  are distant enough, but not too much, from
  autarky prices.
• Terms of trade effect corresponding to distance
  between autarky and trade equilibrium contribute
  to increasing social welfare
• Asymmetric wage effects and the efficiency cost
  of correcting for them contribute to reducing
  social welfare.
• See also Guesnerie (1998, 2002), Feenstra (1987),
  Dixit and Norman (1986),

43
5. Poverty and trade the key relationship
between trade and growth

• Absolute vs. relative inequality trade may be
  changing the relative income of groups of
  individuals but this may matter little if all
  income actually grow implying that poor become
  less poor
• Statics vs. Dynamics all what precedes is about
  the static effects of trade. However, trade may
  affect individual incomes dynamically through the
  rate and the structure of growth of the economy
• Hence the idea of considering in this section the
  relationship between trade, or more exactly trade
  policy, and poverty, which actually is a
  relationship between trade, growth and
  distribution.
• This section based on Winters, McCulloch and
  McKay (2004)

44
Channels through which trade liberalization (TL)
may affect poverty

• Growth and productivity Does TL foster growth
  and poverty reduction? Through what channel
  (accumulation vs. productivity)?
• Households and markets Are border price changes
  passed through onto (poor) households? How are
  they affected? Is the structure of markets
  modified? What is the capacity of response of
  (poor) households?
• Wage and employment Does TL increase wages and
  employment among poor people? Are transitional
  costs born mainly by the poor?
• Macro-stability and uncertainty Does TL increase
  macro-economic instability? Does it create more
  vulnerability to poverty or is vulnerability
  affected directly through micro channels?
• Government revenues and public spending Does TL
  reduce governement revenues and public spending
  in favor of the poor?
• This section offers a brief synthesis of the
  evidence available on these various questions

45
5.1 The growth channel

• Link growth-poverty reduction is very strong
  (Ravallion, 2001, Bourguignon, 2002). Growth
  probably the strongest channel linking trade and
  poverty reduction.
• Trade and growth old debate in economics
• Theory
• Growth boosted by TL, due to
• access to technology and efficient intermediate
  and capital goods
• economies of scale
• competition
• flexibility of a more 'marketed' economy.
• But negative effects too specializing on raw
  materials and extractive industry, destruction of
  infant industry, ...
• ? trade and growth empirical issue

46

• Empirics
• Simple-minded cross-country regressions suggested
  a strong relationship between trade and growth
  (Sachs and Warner, 1995, Dollar and Kraay, 2001,
  Edwards, 1998, Wacziarg and Welch, 2003)
• But weak empirical framework (Rodriguez and
  Rodrik, 2001).
• Problems measuring "trade openness" causation
  doubtful because of endogeneity problems,
  instruments not always convincing TL policy
  packages include other (complementary)
  instruments or TL correlated to other policies.
• Case studies TL often as preliminary to growth
  acceleration
• Structural models

47
Trade, growth synthesis

• Little doubt that trade restrictions are harmful
  to growth ( no evidence of trade restriction
  fostering growth)
• However, doubts about the nature of policies
  leading to that result
• Many fast-growing open economies started with
  strong (subsidized) export sector and high
  protection they liberalized imports in a second
  stage
• Case studies suggest liberalization of imports
  promotes growth only selectively
• Few examples of unilateral liberalization leading
  to faster growth (Chile?)

48
5.2 Trade and productivity gains

• Productivity gains necessary for growth (but not
  sufficient, e.g. fall in inputs).
• Same ambiguity arises from theory as for growth.
• Empirically
• cross-country evidence of impact of RD
  accumulated in imports of intermediate and
  capital goods on TFP growth (Coe, Helpman and
  Hoffmaister, 1997)
• Cross-sectoral studies mixed results, trade not
  necessarily the major force
• Firm-level analysis Bigsten et al. (2000) find
  exports fosters productivity in Africa whereas
  little evidence found for Asia and Latin-America
  (Tybout and Westbrook, 1995). (Problem of
  selection through firm demographics)
• Most studies refer to industrial sectors and
  firms little evidence in other sectors
  (agriculture)
• Ambiguity arising from measurement of TFP
• Link to poverty may be somewhat unclear (input
  effect TFP gains limited to sectors of less
  concern for poor people)

49
5.3 Households and markets the key relationship
between trade and household welfare

• What is the direct impact of trade openness at
  the micro-economic household level?
• Basic mechanism in micro-studies of the impact of
  trade on poverty (and distribution, see Sahn and
  Younger, 2003, Ravallion, 2008, Bourguignon and
  Spadaro, 2006)
• Important issues include
• the impact of trade policies on consumer or
  producer prices
• the structure of the markets where goods i are
  exchanged (markets may not exist, goods may not
  be exchanged and vice-versa)
• the household response to price changes matters
  too.

50
Transmission of Border-price changes

• Relationship between border prices and domestic
  prices depends on transport costs and other costs
  of distribution, competition between traders
  (marketing boards), infrastructure, domestic
  taxes and regulation.
• Example
• Passthrough is high in the absence of transaction
  costs or exchange rate reaction, and with
  competitive traders
• Passthrough may be very low in the opposite
  cases example of prices in China as a function
  of the distance from import and export centres.
• Most available evidence for agricultural products
  point to a diversity of situations

51
Creating and destroying markets

• Some goods may be non-traded because of the
  absence of a market, or high transaction costs.
  In agricultural households, it is then observed
  that qi ci .
• If a market is created, possibly following trade
  liberalization, impact on household welfare may
  be big. Relationship between border prices and
  domestic prices depends on transport costs and
  other costs of distribution, competition between
  traders (marketing boards), infrastructure,
  domestic taxes and regulation.
• Example
• Privatization of marketing board for exports
  makes inputs unaivalable to small farmers who go
  back to subsistance
• Construction of infrastructure reduces
  transaction costs and opens new markets
• The 'household responsibility system' in China as
  market creation
• Exports of rice and coffee as market creation or
  market expansion in Vietnam

52
Household response

• In previous cases, difficult to stick to the
  marginalist framework described by ?pi. Big
  changes are to be expected in production and/or
  consumption, depending on the capacity of
  households to respond to changes in incentives,
  and in particular market structures.
• Note also that secondary effects are to be
  expected from household response that can offset
  or invert initial impact of trade liberalization.
• Examples
• Households with little assets or wealth less
  likely to react to market creation or price
  changes. Liberalization must come with other
  policies so as to benefit the poor.
• NAFTA reduced the price of corn in Mexico, yet
  farmers increased planted areas for lack of good
  substitute or because price of substitutes fell
  too. ( constant price of 'tortilla')
• Problem of incoporating empirical demand
  functions in the analysis

53
5.4 Other channels of the impact of trade on
poverty

• Wage and employment (See section 2 above,
  however take into account labor market
  imperfections). Issue of the transitional periods
  and vulnerability of the poor .
• Macro instability and poverty vulnerability It
  turns out that ToT variability is a minor source
  of macro instability if accompanied by the
  appropriate macro policy (partial adjustment
  strategy). Policy more than ToT shocks seem
  responsible for instability Loayza and
  Hnatkovska (2006).
• Poverty vulnerability is reinforced by trade
  liberalization with high pass-through in
  comparison with subsistance farming (portfolio
  choice). At the same time expected income is
  higher, provided there is no poverty trap. But
  limited evidence of such traps. Extreme
  importance of credit market (and micro-credit).
  (Morduch, 20 )
• Government revenue and spending. Trade does not
  seem to be the main source of variability but
  rather macro policies.

54
Conclusion on trade and poverty

• Growth/productivity as the main and less
  ambiguous channel through which trade and trade
  policies may affect poverty
• Other channels show that complementary policies
  necessary to reinforce the direct impact of trade
  policies on poverty (micro-credit to minimize
  vulnerability, macro-policies to avoid
  instability, regulation to avoid marketing firms
  to capture gains, infrastructure investment to
  facilitate integration in trade flows,
• Not clear, however, that, as such, trade policies
  are superior to other more direct policies in
  alleviating poverty

55
General conclusions on globalization through
trade and equity

• Globalization through trade and equity
• International inequality fell because of trade
  globalization (through the growth channel)
• But not clear that inter-country inequality fell
  too, at least until the early 2000s) some
  countries were not able to integrate into world
  trade flows.
• Inequality within country has probably been
  affected by increased trade, but this effect is
  difficult to measure and many other factors may
  explain the change in inequality
• Some effects of globalization are voluntary
  (trade policies) others are involuntary (changes
  in world economy). This makes other policies all
  the more important to tackle the consequences of
  globalization
• More generally, importance of policies that
  'accompany' trade or directly address
  distributional issues.

56
References

• Acemoglu, 2002, Technical Change, Inequality, and
  the Labor Market, Journal of Economic Literature,
  vol. 40(1), pages 7-72,
• Autor, Levy, Murnane, 2003, The Skill Content Of
  Recent Technological Change An Empirical
  Exploration, The Quarterly Journal of Economics,
  vol. 118(4), 1279-1333
• Atkinson, A. (2000),
• Balassa (1986), The employment effects of trade
  in manufactured products between developed and
  developinfg countries, Journal of Policy
  Modeling, 8(3), 371-90
• Berman, Bound and Griliches, 1994, Changes in the
  demand for skilled labor within US manufacturing
  evidence from the annual survey of manufactures,
  The Quarterly Journal of Economics, 367-397
• Bhagwati and Srinivasan, 1998, Lectures on
  international trade, MIT Presss
• Bigsten et al. (2004), Do African manufacturing
  firms learn from exporting? Journal of
  Development Studies, 40(3), 115-41
• Borjas, Freeman and Katz, 1991. "On the labor
  market effects of immigration and trade", NBER
  Working Papers, 3761 .
• Bourguignon and Spadaro, 2006, Microsimulation as
  a tool for evaluating redistribution policies,
  Journal of Economic Inequality, 4(1), 77-106
• Bourguignon, 2002, The Growth Elasticity of
  Poverty Reduction Explaining Heterogeneity
  across Countries and Time Periods, in T. Eicher
  and S. Turnovski (eds), Growth and Inequality,
  MIT Press, 2003, p. 3-26.
• Cline, 1997, Trade and Income Distribution,
  Institute for International Economics, Washington
  DC

57
References

• Coe, Helpman and Hoffmaister, 1997, North-South R
  D Spillovers, The Economic Journal, Vol. 107,
  No. 440, 134-149.
• Dollar and Kraay, 2004, Trade, Growth, and
  Poverty, The Economic Journal, 114(493), F22-F49
• Edwards, 1998, Openness, Productivity and Growth
  What Do We Really Know? The Economic Journal,
  108(447), 383-398
• Dixit and Norman (1986), Gains from trade without
  lump-sum compensation, Journal of International
  Economics, 21(1-2), 111-122
• Freeman, 1995, Are your wages set in Beijing?,
  The Journal of Economic Perspectives, 9(3), 15-32
• Guesnerie, 2001, "Second Best Redistributive
  Policies the case of International Trade",
  Journal of Public Economic Theory, 3 (1), p.15-25
• International Monetary Fund (2007a), Spillovers
  and cycles in the global economy, World Economic
  Outlook, April
• International Monetary Fund (2007b),
  Globalization and Inequality, World Economic
  Outlook, October
• Krueger, 1993, How computers have changed the
  wage structure evidence from microdata,
  1984-1989, The Quarterly Journal of Economics,
  33-60
• Krugman, 1995, Growing World Trade Causes and
  Consequences, Brookings Papers on Economic
  Activity
• Lawrence and Slaughter, 1993, International trade
  and American wages in the 1980s Giant sucking
  sound or small hiccup, Brookings Papers on
  Economic Activity. Microeconomics
• Leamer (1994), Trade, wages and revolving door
  ideas, NBER, 4716

58
References

• Loayza and Hnatkovska (2005), Volatility and
  Growth, Policy Research Paper 3184, The World
  Bank.
• Martin and Evans (1981), Notes on measuring the
  employment displacement effect of trade, Oxford
  Economic Papers
• Mirrlees, 1971, An Exploration in the Theory of
  Optimum Income Taxation", Review of Economic
  Studies,
• Morduch, 1999, The microfinance promise, Journal
  of Economics Literature, 37, 1569-1614
• Ravallion, 2001, Growth, Inequality and Poverty
  Looking Beyond Averages, World Development,
• Rodriguez and Rodrik, 2000, Trade and growth a
  skeptic's guide to the cross-national evidence,
  NBER Macroeconomics annual, 261-
• Sachs and Warner, 1995, Economic reform and the
  process of global integration, Brookings Papers
  on Economic Activity, 1-37
• Sahn and Younger, 2002, Estimating the incidence
  of indirect taxes in developing countries, in
  Bourguigon and Pereira (eds), The Impact of
  economic policies on poverty, Oxford University
  Press
• Spector, 2001, Is it possible to redistribute the
  gains from trade using income taxation?, Journal
  of International Economics, 55(2), 441-60
• Stiglitz, 1982, Self-Selection and Pareto
  Efficient Taxation, Journal of Public Economics,
  Vol. 17, No. 2, pp. 213-240. ...

59
References

• Tybout and Westbrook, 1995, Trade liberalization
  and the dimensions of efficiency change in
  Mexican manufacturing industries, Journal of
  International Economics, 39(1-2), 53-78
• Winters, McCulloch, McKay (2004), Trade
  liberalization and poverty the evidence so far,
  Journal of Economic Literature, 42, 72-115
• Wood (1994), North-South trade, employment and
  inequality changing fortune in a skill-driven
  world, Oxford University Press