1

The path to global free trade:Spaghetti bowls as building blocs

Richard E. Baldwin

Professor of International Economics

Graduate Institute of International Studies, Geneva

World Economy Annual Lecture

Nottingham 22 June 2006

2

Introduction & Plan• Topic: Final steps to global free trade.

– Bhagwati’s paths.– Define: global free trade.– Bold prediction & caveats.

• Outline1. Political economy of liberalisation.

2. Structured historical narrative.

3. Staging post 2010.

4. The final 2 steps.

5. WTO’s job.

3

6 stylised facts1. The GATT process started when tariffs were very

high worldwide;2. Rich nations liberalised much more than poor

nations, in both the GATT process (i.e. bound rates) and RTAs;

3. The liberalisation focused on industrial goods in which two-way trade in similar goods is prevalent;

4. The process took 40 years;5. Some sectors were excluded entirely and others

experienced much less tariff cutting;6. Regional tariff cutting went hand-in-hand with

multilateral liberalisation.

4

Political Economy• Status quo tariff.

• Tariff balances supply & demand for protection.

• Supply = marginal cost curve (marginal welfare damage).

• Demand = marginal utility curve (marginal benefit to special interests).

5

Protection demand

quantity

eurosS

D

Pw

TP

a P’P

a P’

T’

Protection demand

euros

T’

6

gfedT

dPM

dT

dMT

dT

dW w

)(

quantity

eurosS

D

Pa

b

c

d

g

P’

Pw’

T’

Pw

fe

euros

Protection Supply

Protection supply

T’

Naïve opt’l tariff

7

Protection demand

Protection Supply (unil.)

euros

Juggernaut: demi-cycle 1• Political equilibrium

tariff balances S & D for protection.

• Reciprocal trade talks re-align political economy forces inside each participating nation.

• Protection supply shifts up.– Exporters become

anti-protectionists.

Protection Supply (recip.)

T’

8

Juggernaut: demi-cycle 2• Politically optimal

tariff depends of size of import competing sector.– Walrasian entry, n.

• Number of import competing firms rise with tariff.

• FE curve.• Same for export

sector, but can’t plot.

Tariff,T

n

To

no

FE

9

Juggernaut effect• Fold supply &

demand into GFOC.– Solution to govt first

order condition.

• GFOC rises with n since politically optimal T rises with n.– More to protect on

margin.

Tariff,T

n

To

no

Eo

FE

GFOC (unil)

10

Juggernaut effect• Start: unilaterally

politically optimal tariff, To.

• Reciprocal trade talks shift GFOC down.– Lower opt’l T for any

given n (exporters are now in the game).

• Juggernaut rolls forward to E’

A

Tariff,T

n

ToEo

FE

GFOC (recip)

GFOC (unil)

E’

T1

T’ B

11

Juggernaut effect• If all exports covered

by the reciprocity, politically optimal tariff is zero.

• Could take decades crush the anti-trade forces (import competing firms) & build pro-trade forces (export firms).

Tariff,T

n

ToEo

FE

GFOC (MTN)

GFOC (unil)

Efinal

12

Domino effect• Demi-cycle I: Idiosyncratic formation or deepening

of a trade bloc re-aligns the political economy forces inside non-member nations.

• Pro-membership political economy forces:– Non-member exporters: Trade diversion (fresh loses) &

Trade Creation (lost opportunity).

• Anti-membership political economy forces:– If deeper, may resist more.

• If export sectors are politically larger than import competing sectors.

• Demi-cycle II: if a new member joins, “forces for inclusion” get stronger in non-member nations.

13

RTB unilateralism• Competition for out-sourcing jobs and investment drive

nations to unilaterally cut tariffs.• Re-aligns political economy forces in DCs.• Unbundling of manufacturing process (i.e.

fragmentation, vertical differentiation, slicing up value added chain) is key.– Destroys import substitution (scale, competition)– Makes export-led industrialisation more successful (foreign

technology from MNC/buyers, ready market).

• Finer division of labour may mean no import competing industry.

• MNC role may imply imports mostly re-exported.– Importers are also exporters => no political economy

conflict.

14

Ancillary effects• Intra-sectoral special interest politics.

– Melitz model & reciprocal liberalisation: big firms win, small firms lose.

– Add Mancur Olsen’s Asymmetry on political organisation & juggernaut effect works well on intra-industry trade.

• Losers lobby harder.• Home market magnification effect.

– Industry becomes more footloose, not less, as trade barriers fall.

– Competition for industry becomes more fierce as tariffs fall globally. (Small preference margins can matter a lot).

15

Historical Narrative• 3 key effects: Jugger (MTNs), Domino (RTAs), RTB

(unilateralism).• “Empirical evidence” intended to “demonstrate”

usefulness of the 3 key effects.• Line sketch.• Can’t pretend to explain everything.

16

Historical Narrative• 1947-1958.

– GATT starts.

– Juggernaut works but stops in 1950s.

0

10

20

30

40

50

60

70

1920

1922

1924

1926

1928

1930

1932

1934

1936

1938

1940

1942

1944

1946

1948

1950

1952

1954

1956

1958

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

%

Geneva Round, '47

Annecy Round, '49

Torquay Round, 51Geneva Round, 56

Kennedy Round, '67Dillon Round, 56

1934 Trade Act

17

Dominos trigger juggernauts• 1958-1972.

– EEC formation:• Europe domino effect

phase I.• Kennedy Round.

• RTAs: US Auto Pact & EEC, EFTA.

• MTNs, RTAs & unilateralism proceed in tandem.

• Liberalisation begets liberalisation.

T

n

FE

GFOC (unil)

FE’ FE”

E”

T

n

FE

GFOC (MTN)

FE’ FE”

E’

Eo

Efinal

18

1973-1985• EEC first

enlargement and EEC-EFTA FTAs, create another incentive for an MTN (Tokyo Round, 73-79).

• Stagflation postpones all forms of trade liberalisation.

0

10

20

30

40

50

60

70

1920

1922

1924

1926

1928

1930

1932

1934

1936

1938

1940

1942

1944

1946

1948

1950

1952

1954

1956

1958

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

%

Geneva Round, '47

Annecy Round, '49

Torquay Round, 51Geneva Round, 56

Kennedy Round, '67Dillon Round, 56

1934 Trade Act

Tokyo

19

1986-1990• Juggernaut

& domino re-engage.

• Single European Act, 1986.

• US-Canada FTA talks start, 1986.

• Uruguay Round starts, 1986.

US & EU average applied industrial tariffs

EU

US

0

1

2

3

4

5

6

7

8

9

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

%

Uruguary Round tariff cuts begin

20

1990-1994• European spaghetti bowl forms.

– USSR collapse.

• North American spaghetti bowls forms.– US-Mexico FTA triggers massive domino effect.– NAFTA, Mercosur, dozens of spoke-spoke FTAs,

long queue for US bilaterals.

21

1994-2000• European spaghetti bowl advances.

– Euro-Meds, etc.

• North American spaghetti bowls advances.– NAFTA crushes Mexican anti-trade forces.– Mexico ‘sells’ its politically optimal tariff cuts in

over 40 bilaterals.• Japan, EU & US.

22

1986-2000• RTB unilateralism in East Asia (circa 1985).

Reductions in applied MFN tariffs on

Asian crisis

23

MTNs, RTAs & unilateralism• In 1990s, as in the 1960s & 1980s, all the ‘isms’

progress hand-in-hand.

• No evidence that ‘isms’ are substitutes.

24

2000-2006• East Asian noodle bowl starts.• China’s approach to ASEAN for FTA triggers

domino effect in East Asia.– ACFTA, AKFTA, AJFTA, many bilaterals.– No significant regionalism to date, but lots

promised.

• Western Hemisphere spaghetti bowl advances.– US opposition to FTAs crushed by NAFTA; US

follows promiscuous FTA strategy.

• European FTAs multiply, spokes start to proliferate FTAs.

25

Staging Post 2010• Europe, North America and East Asia: ‘fuzzy’,

‘leaky’ trade blocs.– North America & Europe done deals; between &

within near-duty-free status (major flows).

• Many East Asian FTAs may have problems (typically south-south), but Japan-Malaysia, & 4 other big ASEANs very likely to be implemented.– Rest due to domino and RTB unilateralism.

• Prediction: Applied tariffs will be near zero for world’s major trade flows around 2010.

26

Fractals• Definition: “A rough or fragmented geometric

shape that can be subdivided in parts, each of which is (at least approximately) a reduced/size copy of the whole.”

• World trade system made of 3 fuzzy, leaky trade blocs each of which is made up of fuzzy leaky sub-blocs.

• The point:

• Solution to one is the solution to all (roughly).

27

PECS• How PECS fixed the European spaghetti bowl

and why.

• Spaghetti bowl problems: – Multi ROOs (hard to do biz in spokes)– Bilateral cumulation (hinders efficient sourcing in

spokes)

• 1997, EU set up PECS: – imposed common set of ROOs on EU, EFTA &

CEECs.– Imposed diagonal cumulation.

28

PECS• Spaghetti bowl is not by accident.

– Pair-specific political economy forces => pair-specific policy; especially hub & spoke.

• Unbundling & off-shoring– Former beneficiaries of complexity downsized and off-shored

from EU.– Some EU firms set up in spokes and are now harmed by the

complexity (“us” becomes “them”).– EU firms push EU to tame the tangle of FTAs.

• “Spaghetti bowl as building blocs”• Complexity & unbundling create new politically

economy force – Push system the short distance from near-free trade with

matrix of bilaterals to free trade ‘lake.’– Multilateralise the FTAs.

• Domino effect in ROOs/Cumulation.

29

2 final steps• Penultimate:

• Multilateralise the spaghetti bowl FTAs in North America (when US firms become victims of the complexity).– NAFTA ROOs already popular.– Diagonal cumulation is next.

• East Asia, much more difficult.– Might be RTB unilateralism makes

ROOs/Cumulation irrelevant. – ASEAN ROOs already popular.

30

2 final steps• Ultimate

• As global unbundling continues, SBBB pressures will mount.

• Impractical to do a PECS given PECS, NAFTA & ASEAN ROOs.

• Alternative is ITA-like agreement with coverage of most industrial goods. – Zero-for-zero.

31

WTO’s role• Penultimate: not much, should study impact of

PECS on non-members more closely.

• The final step could be in WTO, but not like DDA.

• ITA was much more like old-fashioned GATT Rounds. – Much less emphasis on S&D.– Principle supplier ideas.– RTB unilateral to bring newly industrialising nations

along.

32

ENDThanks for listening:

www.hei.unige.ch/baldwin/