Investor-State Investor-State Treaty Regimes Treaty Regimes

and Arbitral and Arbitral Processes Processes Exploratory ResearchExploratory Research

Beth A. SimmonsBeth A. Simmons

Straus Institute, 2009-2010Straus Institute, 2009-2010

Introduction to the TopicIntroduction to the Topic Characteristics of the emerging “regime” for Characteristics of the emerging “regime” for

the facilitation and protection of FDIthe facilitation and protection of FDI Uniqueness of the emerging systemUniqueness of the emerging system

Largely bilateralLargely bilateral Largely North-SouthLargely North-South Special status of investors in dispute settlementSpecial status of investors in dispute settlement

Growing questions of legitimacyGrowing questions of legitimacy Asymmetrical aspectsAsymmetrical aspects Not especially transparentNot especially transparent Law BIT by BIT? Law through litigation? Law BIT by BIT? Law through litigation? Increasingly, state actors are resisting: exit and Increasingly, state actors are resisting: exit and

voicevoice My purpose: to develop a theory of the way My purpose: to develop a theory of the way

this system worksthis system works

0

10

20

30

40

50

60

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008

WTO Trade Disputes and ICSID Investment Disputes

number of new ICSID arbitrations number of new GATT/WTO disputes

Introduction to the TopicIntroduction to the Topic Characteristics of the emerging “regime” for Characteristics of the emerging “regime” for

the facilitation and protection of FDIthe facilitation and protection of FDI Uniqueness of the emerging systemUniqueness of the emerging system

Largely bilateralLargely bilateral Largely North-SouthLargely North-South Special status of investors in dispute settlementSpecial status of investors in dispute settlement

Growing questions of legitimacyGrowing questions of legitimacy Asymmetrical aspectsAsymmetrical aspects Not especially transparentNot especially transparent Law BIT by BIT? Law through litigation? Law BIT by BIT? Law through litigation? Increasingly, state actors are resisting: exit and Increasingly, state actors are resisting: exit and

voicevoice My purpose: to develop a theory of the way My purpose: to develop a theory of the way

this system worksthis system works

Comparison of defendants Comparison of defendants in trade and investment in trade and investment

disputesdisputes

References to "ICSID", UNCITRAL" and "International Arbitration" in U.S. news sources, 1990-2009

0

10

20

30

40

50

60

70

80

90

ICS

ID a

nd

UN

CIT

RA

L h

its

0

50

100

150

200

250

300

350

400

450

500

Inte

rnat

ion

al A

rbit

rati

on

hit

s

"ICSID" hits "UNCITRAL" hits "International Arbitration" hits

Source: LexisNexis database

References to "ICSID, "UNCITRAL" and "International Arbitration" in world news sources, 1990-2009

0

20

40

60

80

100

120

140

160

180

200

ICS

ID a

nd

UN

CIT

RA

L h

its

0

200

400

600

800

1000

1200

1400

Inte

rnat

ion

al A

rbit

rati

on

hit

s

ICSID hits UNCITRAL hits Int. Arbitration hits

Source: LexisNexis database

Outline of presentationOutline of presentation

The rise of BITsThe rise of BITs BITs as bargains – a look at contentsBITs as bargains – a look at contents The consequences of BITsThe consequences of BITs Trends in arbitrationTrends in arbitration Push-back: annulment proceedingsPush-back: annulment proceedings ConclusionsConclusions

The Rise of BITsThe Rise of BITsCumulative BITs world wide

0

500

1000

1500

2000

2500

3000

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

Average number of BITs negotiated each Average number of BITs negotiated each year by…year by…

0

0.5

1

1.5

2

2.5

3

3.5

left governments center government right governments

Average, left = 1.45/year**Average, center = 1.72/yearAverage, right = 1.67/year*

Why the BIT mania?Why the BIT mania?

Technology for joint gainsTechnology for joint gains Spread of neo-liberal ideas of Spread of neo-liberal ideas of

developmentdevelopment

But alsoBut also High stakesHigh stakes

Tough competitionTough competition Hard timesHard times

1

2

3

4A

vera

ge F

ore

ign

Dir

ect

Inv

est

men

t, N

et

Infl

ow

s (%

of

GD

P)

0

50

100

150

200

Nu

mb

er

of

Tre

ati

es

Sig

ned

1960 1970 1980 1990 2000

Year

Average Foreign Direct Investment

Number of Treaties Signed

High stakesHigh stakes

Tough CompetitionTough Competition

Predictors that a developing will ratify a Predictors that a developing will ratify a BIT with any other country:BIT with any other country: Ratification in countries with similar Ratification in countries with similar

infrastructureinfrastructure Ratification in countries with similar work Ratification in countries with similar work

forceforce Ratification in countries with similar export Ratification in countries with similar export

profileprofile Developing countries do what their closest Developing countries do what their closest

economic competitors for capital do.economic competitors for capital do.

Hard Hard TimesTimesWorld GDP Growth and BITs

0

1000

2000

3000

4000

5000

6000

1980 1985 1990 1995 2000 2005

Year

Nu

mb

er o

f tr

eati

es s

ign

ed

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

Wo

rld

GD

P g

row

th

cumulative BITs world GDP growth 5 per. Mov. Avg. (world GDP growth)

Average Growth Rates Preceding Any BIT Signing

0

0.5

1

1.5

2

2.5

3

3.5

4

-4 -3 -2 -1 0 1 2 3 4

Year (0=year of BIT signing)

gro

wth

rat

e

Any BIT signed No BITsigned

Note: difference in mean growth rates is statistically significant (p<.05) in years -3 and -2.

Average Growth Rates Preceding BITs Sprees (>4 BITs/year)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

-4 -3 -2 -1 0 1 2 3 4

Year (0=BITs spree year)

gro

wth

rat

e

BIT spree no BIT spree

Note: difference in mean growth rates is statistically significant (p<.05) in years -3,-2,-1 and 0.

BITs as bargains: a look BITs as bargains: a look at contentsat contents

Getting the text – not as easy as it sounds.Getting the text – not as easy as it sounds. Deciding what to code – not as obvious as it Deciding what to code – not as obvious as it

may seem.may seem. Starting point: dispute settlement clauses.Starting point: dispute settlement clauses. Simple assumptions:Simple assumptions:

Capital importing countries want as much local Capital importing countries want as much local control as they can get away withcontrol as they can get away with

Investors want a dispute as removed from local Investors want a dispute as removed from local host country institutions as possiblehost country institutions as possible

Hypothesis: host countries in weak Hypothesis: host countries in weak bargaining positions lose more sovereign bargaining positions lose more sovereign control in the event of disputescontrol in the event of disputes

Hard-to-find BITs

0

50

100

150

200

250

1959

1962

1965

1968

1971

1974

1977

1980

1983

1986

1989

1992

1995

1998

2001

2004

year

nu

mb

er o

f n

ew B

ITs

texts found texts missing

"missing" = could not be found on UNCTAD or major country-specific websitesSource: Allee and Peinhardt, 2010

How random are the “hard-to-How random are the “hard-to-find” BITs?find” BITs?

-10

-8

-6

-4

-2

0

2

4

6

8

10

poorer partner richer partner combined

(les

s d

em)

po

lity

sca

le (

mo

re d

em)

texts found texts missing

Mean democracy scores of countries' whose BITs texts are "easy" versus "difficult" to find are significantly different (p=.0000)."Found" BIT texts are likely to be a biased (highly democratic) sample.

BITs as bargains: a look BITs as bargains: a look at contentsat contents

Getting the text – not as easy as it sounds.Getting the text – not as easy as it sounds. Deciding what to code – not as obvious as it Deciding what to code – not as obvious as it

may seem.may seem. Starting point: dispute settlement clauses.Starting point: dispute settlement clauses. Simple assumptions:Simple assumptions:

Capital importing countries want as much local Capital importing countries want as much local control as they can get away withcontrol as they can get away with

Investors want a dispute as removed from local Investors want a dispute as removed from local host country institutions as possiblehost country institutions as possible

Hypothesis: host countries in weak Hypothesis: host countries in weak bargaining positions lose more sovereign bargaining positions lose more sovereign control in the event of disputescontrol in the event of disputes

Number of New BITs that Mention ICSID

0

20

40

60

80

100

120

140

year

no mention of ICSID mentions ICSID

Numbert of new BITs that Mention UNCITRAL

0

20

40

60

80

100

120

140

no mention of UNCITRAL mentions UNCITRAL

BITs as bargains: a look BITs as bargains: a look at contentsat contents

Getting the text – not as easy as it sounds.Getting the text – not as easy as it sounds. Deciding what to code – not as obvious as it Deciding what to code – not as obvious as it

may seem.may seem. Starting point: dispute settlement clauses.Starting point: dispute settlement clauses. Simple assumptions:Simple assumptions:

Capital importing countries want as much local Capital importing countries want as much local control as they can get away withcontrol as they can get away with

Investors want a dispute as removed from local Investors want a dispute as removed from local host country institutions as possiblehost country institutions as possible

Hypothesis: host countries in weak Hypothesis: host countries in weak bargaining positions lose more sovereign bargaining positions lose more sovereign control in the event of disputescontrol in the event of disputes

What explains the choice of dispute What explains the choice of dispute settlement provisions?settlement provisions?

Results of a series of probit models; Results of a series of probit models; dependent variable: presence or absence of particular kind of dependent variable: presence or absence of particular kind of

clauseclauseICSIDprovisio

n

UNCITRAL

provision

NeitherICSID

norUNCITRA

L

Local body

provision

ExhaustLocalremedie

s

Growth past

3-yrs (host)

-.005(p=.422

)

-.021***

(p=.003)

.027**(p=.012)

.024***

(p=.000)

.028**

(p=.028)

Jointdemocrac

y

.038***

(p=.000)

.011(p=.112)

-.034***

(p=.000)

-.005(p=.315

)

-.007(p=.477

)

Difference in

development

.096(p=.102

)

-.054(p=.427)

-.078(p=.261)

-.283***

(p=.000)

-.203**

(p=.015)

Year .061***

(p=.000)

.093***

(p=.000)

-.061***

(p=.000)

.050***

(p=.001)

-.037*(p=.098

)

# obs. 1213 1213 1213 1204 1205

Pseudo R2 .120 .091 .135 .082 .053

Data on provisions: Allee and Peinhardt 2010.

Effect of Growth Rate on probability of no ICSID or UNCITRAL clauses

0

0.02

0.04

0.06

0.08

0.1

10% grow th -10% grow th

pro

bab

ility

Effect of Growth Rate on probability of an UNCITRAL clause

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

10% growth -10% growth

pro

bab

ility

Effect of Growth Rate on probability of clause relating to local recourse

0

0.2

0.4

0.6

0.8

10% growth -10% growth

The The consequences of consequences of the host business the host business

cycle for BITs cycle for BITs dispute dispute

settlement settlement clauses…clauses…

What are the consequences What are the consequences of signing BITs?of signing BITs?

Great source of collected articles: Great source of collected articles: Sauvant, Karl P., and Lisa E. Sachs. 2009. Sauvant, Karl P., and Lisa E. Sachs. 2009. The Effect of The Effect of

Treaties on Foreign Direct InvestmentTreaties on Foreign Direct Investment. Oxford: OUP. Oxford: OUP

Effect of BITs signings/ratifications on FDI flows:Effect of BITs signings/ratifications on FDI flows: very little if any, conditional on: very little if any, conditional on:

good domestic institutions (Hallward-Driemeier 2003; good domestic institutions (Hallward-Driemeier 2003; Rose-Ackerman and Tobin 2005) and Rose-Ackerman and Tobin 2005) and

the capital source (Gallagher and Birch 2006).the capital source (Gallagher and Birch 2006). Some positive benefits Eggera and Pfaffermayr 2004; Some positive benefits Eggera and Pfaffermayr 2004;

Neumayer and Spess 2004; Kerner 2009).Neumayer and Spess 2004; Kerner 2009).

The The most predictablemost predictable outcome of ratifying BITs… outcome of ratifying BITs…

BITs: If you Ratify, They Will BITs: If you Ratify, They Will Litigate…Litigate…

BITs and ICSID cases, world wide

0

5

10

15

20

25

30

35

40

1980 1985 1990 1995 2000 2005

year

nu

mb

er o

f IC

SID

cas

es

0

1000

2000

3000

4000

5000

6000

cum

ula

tive

BIT

s

number of new cases cumulative BITs world wide

BITs: If you Ratify, They Will BITs: If you Ratify, They Will Litigate…Litigate…

BITs and ICSID cases, Latin America

0

5

10

15

20

25

1980 1985 1990 1995 2000 2005

year

nu

mb

er o

f IC

SID

cas

es

0

100

200

300

400

500

600

700

cum

ula

tive

BIT

s

Number of Latin American cases cumulative BITs, L.A.

BITs: If you Ratify, They Will BITs: If you Ratify, They Will Litigate…Litigate…

BITs and ICSID cases, Argentina

0

2

4

6

8

10

12

14

16

18

1980 1985 1990 1995 2000 2005

year

nu

mb

er o

f IC

SID

cas

es

0

10

20

30

40

50

60

cum

ula

tive

BIT

s

Number of Argentine cases cumulative BITs, Argentina

Growth in Investor-State Arbitration, ICSID Cases

0

5

10

15

20

25

30

35

40

Year

number of new cases

Average number of Average number of arbitrations by…arbitrations by…

0

0.1

0.2

0.3

0.4

0.5

0.6

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

left governments center governments right governments

Average number of Average number of arbitrations by…arbitrations by…

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

left govs center govs right govs

excluding Argentina

Economic Conditions prior Economic Conditions prior to Arbitrationsto Arbitrations

World GDP Growth and Arbitration

0

5

10

15

20

25

30

35

40

1976 1981 1986 1991 1996 2001 2006

Year

nu

mb

er o

f n

ew c

ases

0

1

2

3

4

5

6

7

Wo

rld

GD

P g

row

th

5 per. Mov. Avg. (number of new cases) 5 per. Mov. Avg. (world GDP growth)

Economic Conditions prior Economic Conditions prior to Arbitrationsto Arbitrations

Latin America & Caribbean GDP Growth and Arbitration

0

5

10

15

20

25

1972 1977 1982 1987 1992 1997 2002 2007

year

nu

mb

er o

f n

ew c

ases

-4

-2

0

2

4

6

8

10

Lat

in A

mer

ica

&

Car

ibb

ean

GD

P g

row

th

5 per. Mov. Avg. (Number of Latin American cases)

5 per. Mov. Avg. (Latin America GDP Growth)

Economic Conditions prior Economic Conditions prior to Arbitrationsto Arbitrations

Arbitration and GDP Growth - Argentina(5-year moving averages)

0

2

4

6

8

10

12

14

16

18

1972 1977 1982 1987 1992 1997 2002 2007

year

nu

mb

er o

f n

ew c

ases

-15

-10

-5

0

5

10

15

5 per. Mov. Avg. (Number of Argentine cases) 5 per. Mov. Avg. (Argentine GDP growth)

Conditions associated with Conditions associated with arbitrationarbitration

Dependent variable: arbitration in a given year (probit Dependent variable: arbitration in a given year (probit coefficients, p-values)coefficients, p-values)Explanatory Variables:Explanatory Variables: Model 2Model 2 Model 3Model 3 Model 4Model 4 Model 5Model 5

Log of arbitration, (t-1)Log of arbitration, (t-1) .202***.202***(p=.000)(p=.000)

.215***.215***(p=.001)(p=.001)

.224***.224***(p=.001)(p=.001)

.220***.220***(p=.000)(p=.000)

log of cumulative # of log of cumulative # of BITsBITs

.133***.133***(p=.000)(p=.000)

.255***.255***(p=.000)(p=.000)

.119**.119**(p=.011)(p=.011)

.224***.224***(p=.000)(p=.000)

YearYear .029***.029***(p=.000)(p=.000)

.024***.024***(p=.001)(p=.001)

.030***.030***(p=.000)(p=.000)

.025***.025***(p=.000)(p=.000)

Log inflation (t-2)Log inflation (t-2) .067*.067*(p=.055)(p=.055)

---- ---- ----

Log change in reserves Log change in reserves (t-2)(t-2)

---- -.088***-.088***(p=.002)(p=.002)

---- ----

FDI outflows (t-2)FDI outflows (t-2) ---- ---- -.004**-.004**(p=.027)(p=.027)

----

Change in foreign debt Change in foreign debt service/GDP(t-3)service/GDP(t-3)

---- ---- ---- -.009**-.009**(p=.047)(p=.047)

Overall R2Overall R2 .074.074 .096.096 .079.079 .092.092

# of countries# of countries 162162 118118 155155 130130

Total # observationsTotal # observations 27392739 20652065 20942094 20372037It’s the economy, stupid!

AnnulmentsAnnulments

No appeals of ICSID tribunal No appeals of ICSID tribunal decisions (compare WTO DSM).decisions (compare WTO DSM).

Annulments on only very narrow Annulments on only very narrow groundsgrounds absence of proper reasoningabsence of proper reasoning manifest "excess of powers" manifest "excess of powers"

And yet…And yet…

ICSID: Awards and annulment registrations

0

2

4

6

8

10

12

14

16

18

awards annulment registrations

In 2008 the number of new registrations for annulment proceedings exceeded the number of awards on the merits* made by ICSID tribunals in that year

*excluding awards on jurisdiction and settlements parties requested to be written in the form of awards

A Closer Look at Annulment A Closer Look at Annulment Attempts…Attempts…

ArgentinaArgentina 88

EgyptEgypt 44

ChileChile 33

EcuadorEcuador 22

MalaysiaMalaysia 22

PeruPeru 22

CameroonCameroon 11

DRCDRC 11

GabonGabon 11

GuineaGuinea 11

GuyanaGuyana 11

IndonesiaIndonesia 11

KazakhstanKazakhstan 11

MoroccoMorocco 11

PhilippinesPhilippines 11

SeychellesSeychelles 11

United Arab United Arab EmiratesEmirates 11

Sectors in which Sectors in which annulments have been annulments have been

sought…sought…

ICSID: Awards and annulment registrations

0

2

4

6

8

10

12

14

16

18

awards annulment registrations

Average democracy score, post 2006: 6/10

Average democracy score, through 2006: 2/10

Things to think aboutThings to think about The model that BITs are freely accepted fully-The model that BITs are freely accepted fully-

informed contracts should be re-examined. informed contracts should be re-examined. Though not coerced, many developing countries Though not coerced, many developing countries

entered these agreements under “stressful” entered these agreements under “stressful” conditions.conditions.

Bargaining asymmetries have consequences:Bargaining asymmetries have consequences: Rapid BIT spreadRapid BIT spread Poor termsPoor terms Insufficient consideration of the risksInsufficient consideration of the risks Systemically – result is an asymmetrical regime that gives Systemically – result is an asymmetrical regime that gives

investors much more than was “customary” and puts most investors much more than was “customary” and puts most of the risk of economic vulnerability on states.of the risk of economic vulnerability on states.

Many are questioning the payoffs (attracting more Many are questioning the payoffs (attracting more litigation than investment?)litigation than investment?)

Fortunately, there are pressures for change…Fortunately, there are pressures for change…

United States Net FDI

-200

-150

-100

-50

0

50

100

150

200

year

bil

lio

ns

of

US

do

llar

s

Complainants and Forums: Canadian, UNCITRAL: ICSID: Mexican, UNCITRAL:

Reasonable responses Reasonable responses

Reduce the asymmetries.Reduce the asymmetries.

Interpret contracts flexiblyInterpret contracts flexibly

Allow for appeals.Allow for appeals.

Rely more on an insurance Rely more on an insurance regime than a litigation regime than a litigation regime. regime.