ECON2915 Economic Growth Lecture 11 : The Heckscher-Ohlin Model. - - PowerPoint PPT Presentation

ECON2915 Economic Growth

Lecture 11 : The Heckscher-Ohlin Model. Andreas Moxnes

University of Oslo

Fall 2016

1 / 36

Introduction

Recall, in specific-factors model: Two products (e.g. agriculture and manufacturing) and two inputs (e.g. labor and capital). One factor is immobile − → Return on immobile factor (e.g. capital return r) not equal across industries. But in the long run, most inputs are mobile. Differences in returns r means that capital will reallocate. E.g. investment in one sector and capital depreciation in other sector. This lecture: Analyze 2x2x2 model (countries, factors, products). All factors mobile. Enables us to analyze long-run

◮ Effect of trade on GDP ◮ Who gains/loses from trade. ◮ 3 important (and classic) trade theorems. 2 / 36

Four facts

1 Since the 1970’s, the US economy has more than doubled. Along with

this, labor productivity grew.

2 Over the same period, wages started to lag behind productivity growth. 3 Inequality in wages also increased; median wages have been stagnant,

and below-median male wages have fallen.

4 This coincides with the big wave of globalization.

Did globalization cause this? We can use the H-O model to analyze the theoretical mechanisms.

3 / 36

Facts 1 & 2

(1992=100)

4 / 36

Fact 3

Real wages for U.S. men by decile:

5 / 36

Today

6 / 36

Norway : Income inequality

After-tax household income, all individuals except students. Share of income going to the x’th decile (1986=100):

80 90 100 110 120 130 140 150 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 10 20 30 40 50 60 70 80 90 100

7 / 36

Norway : Median real income

After-tax household income:

80 90 100 110 120 130 140 150 160 170 180 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

8 / 36

The Model

Comparative-advantage model in which trade is driven by differences in factor endowments across countries. Let’s assume there are only two countries: US and China. Two goods: Apparel (QA) and Plastics (QP). Two factors of production: Skilled and unskilled labor.

9 / 36

The Model

First, we’ll go through the model with fixed-coefficients production (Leontieff) because it’s easier to understand that way. Later, we’ll go through it with general CRS production functions. Throughout, we assume that each country’s factor supplies are fixed;

◮ Each factor is mobile within its country (no specific factors);

All agents take prices as given.

10 / 36

Factor intensities

Produce 1 unit of apparel:

◮ Requires 1 unit of skilled labor (aSA) and 2 units of unskilled labor

(aUA).

Produce 1 unit of plastics:

◮ Requires 3 units of skilled (aSP) and 3 units of unskilled labor (aUP).

Which product is skilled-labor intensive?

11 / 36

Factor endowments

Assume that the US has 72 million unskilled workers (LU), 60 million skilled workers (LS). Assume that China has 540 million unskilled workers, 300 million skilled workers. − → Which country is skilled-labor abundant (unskilled-labor scarce).

12 / 36

Supply : US

Denote output QA and QP. Unskilled labor market clearing: 2QA +3QP = LU = 72 million Skilled labor market clearing: QA +3QP = LS = 60 million Two equations and two unknowns.

13 / 36

Supply : US

Each curve gives combination of (P,A) where all unskilled (skilled) labor is employed. Only one allocation of (P,A) that gives full employment.

14 / 36

Supply : The Math

We have 2QA +3QP = LU QA +3QP = LS Can solve both equations for QP and set them equal to each other: LU 3 − 2 3QA = LS 3 − 1 3QA 1 3QA = LU 3 − LS 3 QA = LU −LS and QP = LS 3 − 1 3QA = 1 3 (2LS −LU)

15 / 36

Relative supply : US

Relative supply QA/QP does not depend on PA/PP.

16 / 36

Relative supply : US

Suppose we doubled the endowment of both kinds of labor – the relative supply of apparel is unchanged. Therefore, we can think of RS as a function of LU/LS alone. Is it increasing or decreasing in LU/LS? I.e., what happens if we raise LU without changing LS?

17 / 36

Relative supply : US

LU ↑ − → Unskilled labor constraint shifts out − → QA up, QP down.

18 / 36

The Rybczynski theorem

QA ↑, QP ↓ when LU ↑. This is a general result.

Theorem

An increase in the endowment of an input leads to increased output in the sector that uses the input intensively and a fall in output in the other sector (holding output prices constant). Intuition: If apparel output ↑, then that sector needs more unskilled & skilled workers. The skilled workers must come from the plastics industry − → QP must go down. Empirical example: Immigration to Norway last decade.

19 / 36

Immigration to Norway

0.05 ¡ 0.06 ¡ 0.07 ¡ 0.08 ¡ 0.09 ¡ 0.1 ¡ 0.11 ¡ 0.12 ¡ 0.13 ¡ 0.14 ¡ 1998 ¡ 1999 ¡ 2000 ¡ 2001 ¡ 2002 ¡ 2003 ¡ 2004 ¡ 2005 ¡ 2006 ¡ 2007 ¡ 2008 ¡ 2009 ¡ 2010 ¡ 2011 ¡ 2012 ¡ 2013 ¡ Foreign ¡born ¡rela7ve ¡to ¡total ¡popula7on ¡ Construc7on, ¡output ¡rela7ve ¡to ¡Mainland ¡GDP ¡

20 / 36

Relative supply and equilibrium

Assume identical rel. demand in China and US → Same RD curve. RS curves:

◮ High rel. supply QA/QP in China (bc LU/LS high). ◮ Low rel. supply QA/QP in US (bc LU/LS low). 21 / 36

The Heckscher-Ohlin theorem

China: Low autarky price PA/PP. US: High autarky price PA/PP. Free trade: Price must be somewhere in the middle.

◮ China exports A, US exports P.

Theorem

Countries export the good that is intensive in the factor in which it is abundant.

22 / 36

Gains from trade in the aggregate

Budget lines must intersect at A bc production (=income) unchanged.

23 / 36

Gains from trade in the aggregate

New budget line cuts through the autarky indiff. curve.

24 / 36

The distribution of income

In a perfectly competitive economy, profits are zero: Apparel: 2wU +wS = PA Plastics: 3wU +3wS = PP We’re interested in real wages, so divide by output prices and solve for wS/PA: Apparel: wS PA = 1−2wU PA (1) Plastics: wS PA = PP 3PA − wU PA (2)

25 / 36

The distribution of income

wU ↑ − → wS ↓ to keep profits constant (at zero). The negative impact is stronger in Apparel bc that industry is unskilled labor intensive. Where lines intersect: Zero profits in both industries.

26 / 36

The distribution of income : The math

Set (1) = (2): 1−2wU PA = PP 3PA − wU PA ⇐ ⇒ wU PA = 1− 1 3 PP PA Higher PP/PA − → lower wU/PA (and even lower wU/PP). Insert the solution back into e.g. (1) to get wS PA = 1−2wU PA = 2 3 PP PA −1 Higher PP/PA − → higher wS/PA (as well as wS/PP). (For US autarky, we get wU/PA = 1−1/(0.48×3) ≈ 0.3 and wS/PA ≈ 0.4).

27 / 36

The Stolper-Samuelson Theorem

In sum: US: Price of unskilled intensive good ↓ − → real wage of unskilled workers ↓, of skilled workers ↑. China: ?

Theorem

A rise in the relative price of a good will lead to a rise in the real return to that factor which is used most intensively in the production of the good, and conversely, to a fall in the real return to the other factor. Combined with the Heckscher-Ohlin theorem, the following must hold: The abundant factor will gain from trade, the scarce factor will lose.

28 / 36

The Stolper-Samuelson Theorem : In pictures

Free trade: PP/PA ↑ − → Plastics curve shifts out (see equation 2) − → real skilled wages ↑. Intuition: PP/PA ↑ generates profits in the Plastics industry − → Wages wSmust go up. But that generates negative profits in Apparel − → wU ↓ .

29 / 36

Appendix: Allowing for substitution

Suppose that apparel and plastics are both produced with CRS production functions using skilled and unskilled labor. Assume that for any ω = wU/wS ratio, the cost-minimizing unskilled-skilled labor ratio for apparel is greater than that for plastics: aUA (ω) aSA (ω) > aUP (ω) aSP (ω) . Apparel is unskilled-labor intensive. Plastics are skill intensive.

30 / 36

Unit isoquants

31 / 36

Zero-profit conditions again

As before, Apparel: aUA (ω)wU +aSA (ω)wS = PA Plastics: aUP (ω)wU +aSP (ω)wS = PP Divide by output prices and solve for wS/PA: Apparel: wS PA = 1 aSA (ω) − aUA (ω) aSA (ω) wU PA Plastics: wS PA = PP aSP (ω)PA − aUP (ω) aSP (ω) wU PA

32 / 36

Stolper-Samuelson again

33 / 36

Stolper-Samuelson again

If PP ↑ − → E to E

′, wS/PA ↑, wU/PA ↓. 34 / 36

Gains from trade in the aggregate

Previously: Production was fixed, no PPF to speak of. Now: PA/PP ↓ − → Specialization towards Plastic (A − → B).

35 / 36

Four Facts revisited

Recall, (i)-(ii) wage growth < productivity growth, (iii) fall/stagnant wages for low skill, (iv) coincides with globalization. HO model consistent with this, but clearly this is not the only factor.

◮ Skill-biased technical change. ◮ Decline of unions. ◮ Decline of real minimum wages. ◮ Immigration. ◮ Outsourcing/offshoring. 36 / 36