mgi us imbalancing act full report

7/27/2019 MGI US Imbalancing Act Full Report

1/112

MKise G Istitute

The US Imbalancing Act:Can the Current AccountDecit Continue?

June 2007


2/112

2

MKise G Istitute

The McKinsey Global Institute (MGI), ounded in 1990, is McKinsey &

Companys economics research arm. MGIs mission is to help business and

government leaders develop a deeper understanding o the evolution o the

global economy, and provide a act-base that contributes to decision making

on critical management and policy issues.

MGIs research is a unique combination o two disciplines: economics and

management. By integrating these two perspectives, MGI is able to gain

insights into the microeconomic underpinnings o the broad trends shaping

the global economy. MGI has utilized this micro-to-macro approach in

research covering over 15 countries and 28 industry sectors, on topics that

include economic productivity, global economic integration, oshoring, capital

markets, health care, energy, demographics, and consumer demand.

MGIs research is conducted by a group o ull-time MGI Fellows based in

oces in San Francisco, Washington, DC, London, and Shanghai and led

by MGIs director Diana Farrell. MGI project teams also include consultants

drawn rom McKinseys oces around the world, and are supported by

McKinseys network o industry and management experts and worldwide

partners. In addition, MGI teams work with leading economists, including

Nobel laureates and policy experts, who act as advisors to MGI projects.

MGIs research is unded by the partners o McKinsey & Company and not

commissioned by any business, government, or other institution. Further

inormation about MGI and copies o MGIs published reports can be oundat www.mckinsey.com/mgi.


3/112

The US Imbalancing Act:Can the Current AccountDecit Continue?

Diana Farrell

Susan LundAlexander Maasry

Sebastian Roemer

McKinsey Global Institute

June 2007


4/112


5/112

5

Preace

This report is the product o a six-month research project by the McKinsey Global

Institute (MGI). The research builds on MGIs previous research in global capital

markets, and draws on the unique perspectives o our colleagues who work with

nancial institutions and companies in the United States and around the world.

Susan Lund, a senior ellow at MGI based in Washington, DC, worked closely

with me to provide leadership or the project. The team also included Alexander

Maasry and Sebastian Roemer, both MGI ellows. In addition, we would like to

recognize the support given by Tim Beacom, a senior analyst at MGI, and Moira

Soronas, a senior analyst at the McKinsey Knowledge Center.

We have beneted enormously rom the thoughtul contribution o our academic

advisors. Martin N. Baily, an advisor to McKinsey and a senior ellow at the Pe-

terson Institute or International Economics, provided valuable input throughout

the project. Richard Cooper, proessor o international economics at Harvard

University, and Kenneth Rogo, a proessor o public policy and economics at

Harvard University, oered insightul comments on the report. William Cline, a

senior ellow at the Peterson Institute or International Economics, gave helpul

eedback on our economic modeling.

We would also like to thank Janet Bush or her editorial eorts, Rebeca Robboy

or leading external communications, and Deadra Henderson or managing report

production. Sara Larsen, executive assistant, provided the team with supportthroughout the project.


6/112

Our goal in this report is to provide business leaders and policy makers witha act-base and insights into one o the most important issues in the world

economy todaythe growing US current account decit. As with all MGI projects,

this work is independent and has not been commissioned or sponsored by any

business, government, or other institution.

Diana Farrell

Director, McKinsey Global Institute

June 2007

San Francisco


7/112

7

Table o contents

Preace 5

Executive summary 9

1. Understanding the US current account decit 17

2. Could the world und a larger US current account decit? 35

3. Export growth: The key to closing the US decit? 9

. Potential dollar depreciation and impact on trade patterns 1

5. Opportunities and challenges or business and policy

makers 75

Appendices

A. Technical notes 81

B. Methodology or modeling a dollar depreciation and impact

on the current account 89

C. Scenarios or dollar depreciation and impact on US trade

patterns 97

Bibliography 107


8/112

8


9/112

9

Executive summary

Many economists believe that the United States current account decit is on

an unsustainable path.1 Economic alarm bells started ringing in the late 1990s,

and yet the decit has continued to grow unabated.2 In modern times, no large

economy has run a decit o this size or such a prolonged period o time. This

unprecedented situation is now commanding the attention o business leaders,

investors, and policy makers around the world.

When the annual US current account decit reached 3 percent o GDP in 1999,

economists warned that the trend was worrisome. In 200, the annual decit

reached .5 percent o GDP a record $857 billion thus compounding economists

concerns (Exhibit 1). To und its chronic decit, the United States now absorbs a

majority o net capital outfows rom other regions o the world (Exhibit 2). The

total US net oreign debt has swollen to $2.7 trillion, leaving the United States

vulnerable to changes in global investors sentiment. I oreign investors were to

lose their appetite or dollar-denominated assets, US interest rates would prob-

ably rise substantially, at least in the short run, thus restraining overall economic

growth. Many economists and commentators believe a major correctioninvolving

a signicant depreciation o the dollaris looming.

This prompted the McKinseyGlobal Institute (MGI) to consider the US currentaccount decit under two very dierent scenarios over the next ve years: the

decit continuing to expand; and the current account coming into balance. On

one hand, could the world und an ever-growing US decit? On the other, i the

1 For example, see Maurice Obsteld and Kenneth Rogo, The unsustainable US current account

position revisited, 2005; Martin Baily, Dollar Adjustment to Reduce US Imbalance, 2007;William Cline, The United States as a Debtor Nation: Risks and Policy Reorm, 2005.

2 Among others, Catherine Mann in Is the US Trade Decit Sustainable?, 1999.


10/112

10

Eiit 1

THE US CURRENT ACCOUNT DEFICIT REACHED$857 BILLION OR 6.5 PERCENT OF GDP IN 2006

Source: Bureau of Economic Analysis; International Monetary Fund; Global Insight; McKinsey Global Institute GlobalCapital Flows Database

US current account balance$ billion

$ billion

% of GDP

% of GDP

-1,000

-800

-600

-400

-200

0

200

1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

-10

-8

-6

-4

-2

0

2

Eiit 2

THE UNITED STATES ABSORBS MOST OF THE WORLD'S NETCAPITAL FLOWS

* Total net inflows exceed net outflows by $148 billion due to statistical errors and omissions; some of this could reflect "gray market"money as well as the fact that some countries do not report inflows and outflows.

Source: International Monetary Fund; McKinsey Global Institute Global Capital Flows Database

UnitedStates

JapanOtherWesternEurope

Africa EurozoneLatinAmerica

Australia-Pacific

UnitedKingdom

Unattri-butedflows*

MiddleEast

EasternEurope

Russia

%of GDP

5.0 1.8 4.9 0.43.5 -1.6 -2.1 -6.6 -5.1 N/A-8.9 -2.8

Canada

-0.7

OtherAsia

-3.4

Average annual net capital inflows minus net capital outflows, 20012005$ billion

-148-137-119

-96-61-52

-32-20-3

9303235

562


11/112

11

decit were eliminated, what would be the impact on the value o the dollar andon US trade patterns? A number o surprising results emerge that challenge

conventional wisdom.

We nd there is nothing inevitable about a correction in the US current account

decit over the next ve years. It could instead continue to grow, and the world

would have enough capital to und it. At current exchange rates, the United

States could trim the decit slightly by increasing service and manuacturing

exportsbut not enough to reverse its current trajectory. I a large dollar depre-

ciation were to occur, we believe it would more likely be gradual than sudden.

Nonetheless, our analysis illustrates how a very large and rapid dollar deprecia-tion could bring the decit back towards balance with signicantly altered trade

patterns. Irrespective o whether the adjustment process is gradual or rapid,

however, business leaders and policy makers should start considering what a

post-devaluation world would mean or them.

ThE US cUrrEnT accoUnT dEIcIT coUld conTInUE To Grow

Our analysis shows that a correction in the US decit is neither imminent nor

inevitable. Under the current pattern o the US current account, world growth and

exchange rates, the US current account decit would reach $1. trillion in 2012,or 9 percent o GDP . For this to happen, however, the current account surpluses o

other countries would also have to grow suciently large to und the decit.We nd

that under reasonable assumptions, these surpluses would reach $2.1 trillion in

2012, providing the capital required. US net oreign debt would reach percent

o GDP , but the United States would still be able to nance the decit because the

implied net oreign interest payments would remain at less than 1 percent o GDP .

That said, there are several scenarios that could limit growth in the global net

capital outfows necessary to und the US decit. For instance, i China were to

increase domestic consumption and reduce its savings signicantly, net capital

outfows rom Chinacurrently a signicant source o unding or the US de-

citcould all dramatically. In our model, this would leave the US decit equiva-

lent to 90 percent o the remaining global net capital outfows, a gure that is

implausibly high. However, we consider this scenario unlikely. MGIs research on

Chinas consumer market shows that although domestic demand will account or

a greater share o GDP , its expansion will come largely rom growth in incomes

and there will be only a slight all in the national savings rate. 3 Other scenarios,

in isolation, would not limit growth in the US decit.

3 From Made in China to Sold in China: The Rise o the Chinese Urban Consumer, McKinsey

Global Institute, November 200.


12/112

12

Neither will the ability o the United States to und its current account decit dependon the price o oil. Oil imports are the largest single component o the US trade

decit. However, although a higher oil price increases the value o US imports, it

also increases the current account surpluses in oil-exporting economies. This

capital is then recycled into global nancial markets. It doesnt matter whether

these petrodollars are invested in Europe or Asia rather than being invested directly

in the United States; by increasing the capital available in the global nancial

system, they still contribute to unding the US current account decit.

ElIMInaTInG ThE cUrrEnT accoUnT dEIcIT woUld EnTaIl a larGE

dollar dEprEcIaTIon

Although our research shows that the US current account decit could plausibly

continue to grow over the next ve years, the United States cannot continue to

build up oreign liabilities orever. Eventually the decit will need to stabilize, or

even decline, relative to the size o the US economy. A major rebalancing o global

demand and a dollar depreciation o historic proportions would be required or

this to happen over the next ve years.

To balance the US current account by 2012, we nd that the dollar would need to

depreciate by 30 percent rom its January 2007 level. Reducing the decit to 3

percent o GDP , a level that many economists believe to be sustainable,5 would

require a 23 percent depreciation. Only once over the last 35 years has the

dollar depreciated by 30 percent within a ve-year periodin 198588. However,

in 1985 the dollar was at its highest level since 1970, whereas today the dollar

is already below its average post-1970 value, and a urther 30 percent all would

leave it at its lowest level since 1970.

In the absence o a depreciation in the dollar, we nd it likely that the United States

could do little more than slow the growth o the decit. Based on a detailed micro-economic analysis o US exports, our ndings suggest that at current exchange

rates the United States could potentially increase its share o service and manu-

acturing exports to other countries by $25 billion and $20 billion respectively.

However, this would require either much aster GDP growth in the rest o the world

than is currently projected, or the United States to win share o imports rom other

Improving energy eciency in the United States could reduce its oil imports and lower the trade

decit. We calculate that the United States could reduce oil imports by 10 percent by 2012 just

by implementing technologies available today that improve energy productivity. See chapter 3

or more detail.5 For instance, see Ahearne, Alan, William Cline, et al., Global Imbalances: Time or Action, IIE

Policy Brie, Peterson Institute or International Economics, Washington, DC, orthcoming in

2007


13/112

13

countries. Even so, at best this increase would merely hold the US current accountdecit to its current level relative to GDP , or .5 percent.

I the dollar were to depreciate, a 30 percent decline could play out in dierent

ways, since currencies rarely move evenly against all others. We modeled three

depreciation scenarios: rst, an even depreciation against all currencies; second,

a scenario under which Asian currencies adjust the most; and third, a scenario

under which Asian currencies maintain their current value and the adjustment

takes place in Europe, Canada and Mexico, and the rest o the world. While the

eects o these scenarios dier, a number o changes in trade patterns appear

under all three.

aTEr dEprEcIaTIon, a larGE US TradE dEIcIT pErSISTS

parTIcUlarly wITh chIna

Surprisingly, even i the United States were to balance its current account, it would

still continue to run a large trade decit. Under all three depreciation scenarios,

the US trade decit in goods would stand at around $720 billion in 2012only

slightly smaller than it is today. However, this decit on merchandise trade would

be oset by a $30 billion surplus on trade in services and by positive net

oreign-income payments. The latter is due to a turnaround in the US oreign debt

position. I current trends were to continue, US net oreign debt would rise to

$8.1 trillion in 2012. However, i the dollar were to depreciate by 30 percent, the

United States would become a net oreign creditorto the tune o $.8 trillion,

generating $35 billion a year in net interest payments to the United States.

Under all depreciation scenarios, the United States would continue to run a

large bilateral trade decit with China. The bilateral decit was $198 billion in

2005more than one-quarter o the total US trade decit. Even i the dollar

were to depreciate by 5 percent against the yuan (as it does under our seconddepreciation scenario), however, the US trade decit with China would still be

$87 billion. The huge cost advantage that China enjoys in producing goods such

as toys and clothing means that the yuan would need to appreciate by more than

50 percent to eliminate the US trade decit with China.

In contrast, US trade with Canada and Mexicocountries oten overlooked in the

current debatewould improve dramatically. Our model shows that the US trade

balance with NAFTA would swing rom a decit o $109 billion to a surplus o

Our model assumes that the dollar depreciates against all Asian currencies by the sameamount. I the yuan appreciated more than other Asian currencies, some production o low-costgoods could shit, over time, rom China to other countries, such as Vietnam or Cambodia. This

could reduce the bilateral decit with China.


14/112

1

$100 billion or more. This would hit the Canadian and Mexican economies hard,and the United States would have to work with them to ease the transition.

SErvIcES and hIGh-TEch ExporTS Show bIGGEST opporTUnITy or

IMprovEMEnT

Although, with a balanced current account, the United States would still have

a trade decit, in the years ollowing a large dollar depreciation US exports

across a range o products would increase dramatically. US services exports

could increase by 30 percent, or $107 billion, rom todays level. Current US

trade surpluses in nancial services, royalties and licenses, business services,travel, and education would grow dramaticallyparticularly to Europe, the United

Kingdom, Japan, and Canada.

Also receiving a strong boost rom depreciation would be US exports o high-tech

machinery, such as computers and semiconductors, medical devices, electrical

appliances and machinery, oce and telecommunications equipment, and arm

and construction equipment. Todays $13 billion US trade decit in this cat-

egory would turn into a surplus o as much as $51 billion. This growth potential

highlights the importance o continued productivity improvements in a sector

that not only generates exports directly but also enables the United States to

produce other technologically sophisticated products, rom surgical equipment

to computerized arm machinery.

an aGEnda or bUSInESS lEadErS and polIcy MaKErS

Although the US current account decit could possibly be reversed over the next

ve years and spark a major decline in the value o the dollar, we believe that the

adjustment is more likely to be gradual. The world could und a larger decit, and

a reversal need not be immediate. Capital infows into the United States have

grown continuously despite the decit, the war in Iraq, the 9/11 terrorist attacks

and other adverse events. Fundamentally, the US economy is strong and oers

an attractive risk-adjusted return or investors. Nonetheless, business leaders

and policy makers should start planning or the possibility that a large dollar all

might unold more rapidly.

Business leaders would do well to consider how a large decline in the dollar

would aect their income statements and balance sheets and what actions

they can take today to prepare or this possibility. Our research yields detailed

insights into what a post-devaluation world would look like. China, or example,would retain its costs advantage as an export location, but Canada and Mexico


15/112

15

could lose theirs. US companies would see growing oreign demand or manytypes o nancial and business services but, to capture this opportunity, they

would have to acquire the requisite language skills and develop products that

meet oreign standards. US companies producing computers, semiconductors,

medical devices, and construction equipment would benet disproportionately.

Much o the public debate over the current account decit, including concern over

the bilateral trade decit with China, is misplaced. There are other more realistic

options or balancing the current account decit, such as improving the US trade

balance with NAFTA and with other Asian economies and expanding service and

high-tech exports.

The primary policy ocus should be on areas oering the United States oppor-

tunities to improve its trade balance signicantly. Service exports clearly have

signicant growth potential, and trade negotiators should continue to reduce bar-

riers to global trade in services. The United States must retain the competitive

environment necessary or its high-tech sectors to drive innovation and R&D in

order to develop the next generation o cutting-edge products. And policy makers

should recognize that trade with Canada and Mexico is at least as important as

that with Europe and Asia, and consequently they should step up eorts within

NAFTA to enhance the areas competitive advantage.

The ollowing chapters discuss these ndings in more detail. Chapter 1 proles

the US current account decit and explores who is unding it. Chapter 2 assesses

how large the decit would be by 2012 i current trends continued, and explores

whether the world could plausibly und an even larger US decit. In chapter 3

a microeconomic approach is employed to examine the potential or the United

States to increase exports o services and manuactured goods without a change

in the value o the dollar. Chapter explores scenarios or dollar depreciation to

reduce the size o the decit and their implications or trade patterns. Chapter

5 outlines the opportunities and challenges acing business leaders and policy

makers.


16/112

1


17/112

17

The United States record-breaking current account decit has triggered alarm

bells concerning the health o the US economy and the potential risks to the

United States and other countries o a sudden reversal in the capital infows

needed to und it.

Beore exploring whether the decit could continue to grow over the next ve

years and what actors might reverse the trend, we rst assess its components,

what has caused the decit to reach its current size, patterns o oreign invest-

ment in the United States, and the dierent ways in which the US decit is

nanced. Readers who are already acquainted with these topics can move on to

chapter 2.

ThE US cUrrEnT accoUnT dEIcIT

A countrys current account is made up o our components: trade in goods,

trade in services, transer payments, and oreign-income payments. In the United

States the trade decit is the largest and astest-growing part o the overall cur-

rent account decit (Exhibit 1.1). Although a great deal o attention is currently

ocused on the bilateral decit with Chinawhich is indeed the largest decit

with a single countrythe United States runs trade decits with virtually every

region o the world. China accounts or roughly 25 percent o the total decit.

(See Box 1 or more detail on the source o US imports.)

Nations trade in both goods and services. The US trade decit in goods is

enormous. It reached $83 billion in 200, or more than $2 billion per day. Thelargest component o this is imported oil and other mineral uels (Exhibit 1.2). As

1. Understanding the US current

account decit


18/112

18

Eiit 1.1

% ofnominalGDP

-4.5 -4.8 -5.7 -6.3 -6.5-3.8-4.2-3.2-2.4

THE TRADE DEFICIT IS THE LARGEST COMPONENT OF THE CURRENTACCOUNT DEFICIT

Note: Figure may not sum to 100% due to rounding.


5.8

N/A

7.4

13.5

-1.4 -1.7-1.6-1.5-1.7-1.3-0.80.0

Compound annualgrowth rate,19902006

%

$ billion (nominal)

-40-40 -38

-43 -45

-53

-50

-59-51

-64

-69

-82

-86

-84

10 1112

24

29

-27-110

-35

29

-79

1990

-76

45

3

1991

24

-95

56

-50

1992

25

-131

60

1993

17

-164

65

-665

51

-662

28

2003

-527

49

-544

25 37

2002

-473

58

-479

2001

-389

61

-424

2000

-416

72

-450

21

1999

-300

80

Transfers

-344

14

1998

-214

80

-245

4

1997

-140

88

-196

13

1996

-125

85

-189

22

1995

-114-85

76

-172

21

1994

-122

2004

-779

62

-792

2005

-7

-836

72

-8552006

Services

Goods

Income

Eiit 1.2

MINERAL FUELS AND MANUFACTURED GOODS CONTRIBUTE MOSTTO THE US TRADE DEFICIT

* Includes low-value imports, trade not classified elsewhere, and non-monetary gold and coin.

Source: United States International Trade Commission; McKinsey Global Institute analysis

US trade balance, 2006$ billion

4

Chemicals

6

Food &agriculture

-836Goodstrade

72

Services

-764

Tradebalance

280

Other*

299

Manu-facturedgoods

44

-299

122

Vehicles &machines

Mineralfuels

Auto-mobiles

93


19/112

19

the price o oil has risen over the last several years this portion o the US tradedecit has soared. Since 200, growth in the cost o oil imports has accounted

or 5 percent o the deterioration in the US trade decit. The United States

also runs large trade decits in manuactured goodsclothes, plastics, toys,

consumer electronics, and urnitureas well as vehicles and machines.

In contrast the United States has consistently run a surplus in services trade

over the last 15 years (Exhibit 1.3). However, at $72 billion in 200, this surplus

is very small in comparison to the goods trade decit. The positive balance in

services was composed o surpluses in business proessional services, royalties

and licenses, and nancial services, with decits in other categories such as

insurance and transportation services.

The overall US trade balance improved between November 200 and February

2007. Over this our-month period the decit was $1 billion smaller than a year

earlier. This was explained primarily by a smaller bill or mineral uels ($1 billion)

refecting a lower oil price; by contrast, the manuacturing trade decit continued

to grow. The trade decit with China deteriorated by $15 billion compared with a

year earlier but improved with all other regions. It remains to be seen whether or

not these improvements continue throughout 2007. Trade fows can be volatile

on a month-to-month but an increase in the US trade decit in March 2007

compared with year earlier gave grounds or caution.

Eiit 1.3

34

27

13

10

25

18

72

35

Businessservices*

Financialservices

Royaltiesandlicenses

Travel TotalOthertransport

1

Passengerfares

Education

5

Insuranceservices

Tele-communi-cations

US SERVICE TRADE SURPLUS IS MAINLY IN BUSINESS SERVICES,ROYALTIES, AND FINANCIAL SERVICES

US services-trade surplus, 2006$ billion

* Computer and information services, management and consulting services, R&D, operational leasing film andother.

Source: Bureau of Economic Analysis; McKinsey Global Institute analysis


20/112

20

Tse mets et ime

The second component o the current account is transer payments. These

represent transers to and rom the United States o unds that are not being

exchanged or nancial assets or goods. The largest types o transers are remit-

tancesmoney sent home by immigrant workersand tax payments to oreign

governments by workers in the United States. The US decit in net transer

payments has grown steadily larger over time, reaching $8 billion in 200.

The nal component o the current account is income payments on oreign invest-

ments. These include interest on bond and bank deposits, dividends on stocks,

and prots earned by oreign subsidiaries.1 Every country has both oreign-income

payments (money sent abroad on oreign-owned assets within the economy) and

oreign-income receipts (money received rom oreign assets owned abroad).

Traditionally the United States has had positive net oreign-income payments.

However, in 200 the United States experienced negative net income or the rst

time ever (Exhibit 1.).

It is astonishing that the United States has been a net oreign debtor to the world

since 198 but has still managed to earn positive net oreign income until last

1 Income earned by subsidiaries abroad is counted as a net income receipt in the currentaccount, whether or not that prot is repatriated to the parent company. I it is reinvested

abroad, then it also generates a new FDI outfow.

Eiit 1.4

-50

-40

-30

-20

-10

0

10

20

30

40

50

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

-3,500

-2,500

-1,500

-500

500

1,500

2,500

3,500

US net foreign income$ billion

US net foreign-asset position$ billion

Net foreign-asset position

US NET FOREIGN INCOME WAS POSITIVE UNTIL

2006, DESPITE GROWING NET DEBT

Net foreign income

Source: International Monetary Fund; McKinsey Global Institute Global Capital Flows Database; McKinsey GlobalInstitute Cross-Border Investments Database


21/112

21

year. The reason or this highly unusual situation is that the United States hasearned a higher rate o return on its investments abroad than oreigners earn in

the United States. Since 1990 the dierence between the two has averaged 1.3

percent (Exhibit 1.5). However, the true spread is likely to be even larger because

oreign-interest payments do not capture unrealized capital gains on equity and

oreign direct investments (FDI)and these make up a larger share o US invest-

ments abroad than oreign investments in the United States (Exhibit 1.).2

All in all, the United States obtains a very good deal on its oreign borrowing. In

200 US net oreign income nally became negative due to the size o the US

net oreign debtnow at $2.7 trillionand to the declining spread on returns o

US assets versus liabilities. I the current account decit were to persist in the

years ahead, US net oreign income would deteriorate urther as it borrows more

each year to nance the decit.

w us te US uet ut eit?

A nations current account refects the dierence between domestic savings and

investment. Any decit must be unded through borrowing abroad. This shows

up in the capital account in the national balance o payments, which refects

the dierence between capital infows into a country and outfows rom that

2 In What explains the US net income balance? 2007, Alexandra Heath estimates that the real

spread on returns o US oreign assets and liabilities has been .7 percent.

Eiit 1.5

0

2

4

6

8

10

1990 1992 1994 1996 1998 2000 2002 2004 2006

RETURNS ON US FOREIGN ASSETS HAVE EXCEEDED

RETURNS PAID ON US LIABILITIES BY 1.3 PERCENT

* Calculated from foreign income receipts and payments. Does not include appreciation of value of assets orliabilities.

Source: International Monetary Fund; McKinsey Global Institute Global Capital Flows Database; McKinsey GlobalInstitute Cross-Border Investments Database; McKinsey Global Institute analysis

Foreign assets: 5.4%

Average spread,19902006: 1.3%

Average annualreturns, 19902006

Foreign liabilities: 4.1%

US foreign income returns on assets and liabilities*%

Foreign liabilities

Foreign assets

Spread on return


22/112

22

country. In principle the sum o the capital account balance and the currentaccount balance should be zero. However, in practice errors in reporting and

statistical discrepancies in data result in dierences.3 In 200, or instance, the

United States had oreign capital infows o $1,75 billion and capital outfows

o $1,0 billion (Exhibit 1.7). This resulted in a capital account surplus o $721

billion compared with a current account decit o $857 billion. The $13 billion

gap was due to errors and omissions.

To understand which countries are unding the US current account decit, we

consider two dierent perspectives.

eig iestmet i te Uite Sttes

In order to understand who is unding the US decit one must rst look at oreign

investors in the United States. In 2005 they purchased $1.2 trillion o US assets

(equal to US capital infows). This includes oreign direct investment (FDI) into US

companies, purchases o equity, private debt, and government debt securities,

oreign lending to the United States, and deposits into US banks and nancial

institutions (see Box 2 or more detail on capital fows into the United States).

3 Since 2001, the sum o all countries oreign capital infows has exceeded reported capital

outfows by an average o $18 billion per year.

Eiit 1.6

US FOREIGN ASSETS ARE MAINLY IN EQUITY AND FDI,WHILE LIABILITIES ARE IN DEBT AND LENDING/DEPOSITS

* Foreign direct investment.

Note: Figure may not sum to 100% due to rounding.

Source: International Monetary Fund; McKinsey Global Institute Cross-Border Investments Database

US foreign assets and liabilities, 2005%, $ billion

3220

28

17

9

34

30 28

FDI*

100%13,68710,894

Equity

Debt

Lending/deposits

LiabilitiesAssets


23/112

23

The US Bureau o Economic Analysis collects data on the countries that are the

source o these investments. These data do not give an entirely accurate viewo oreign investments, however, because some oreign investors use nancial

intermediaries in the Caribbean, London, and other oshore centers. We develop

a methodology to estimate the true source o these unds as described in

appendix A.

Ater adjusting the data we see that Japan and other Asian nations provide the

largest source o capital infows to the United States, averaging $50 billion per

year rom 200205 (Exhibit 1.8). Over the same period the United Kingdom and

continental European nations together invested just over $00 billion annually.

The remainder o capital infows came rom Canada, the Middle East, and the

rest o the world.

Investors rom dierent regions exhibit preerences or dierent types o US as-

set. While two-thirds o Asian investments are in US government debt, Europeans

invest a larger portion o their unds in US corporate debt and loans and deposits,

while Middle Eastern investors enter US markets through corporate-debt securi-

ties, FDI, and deposits into banks and other nancial institutions such as private

equity unds (Exhibit 1.9).

Most o the money we attribute to the rest o the world was channeled through Caribbeannations. Some o this money may, in act, have originated in the Middle East, Europe, Asia, or

even the United States itsel.

Eiit 1.7

IN 2006, US CAPITAL INFLOWS WERE $1.8 TRILLION,COMPARED TO $1.0 TRILLION OF CAPITAL OUTFLOWS

* Includes foreign direct investment, purchases of equity and debt securities, foreign loans and deposits, andforeign reserve asset purchases.


865798783

740

424

707551

439 427

868

326295383

561504

354413

352

1995 1996

485

1997 1998 1999

1,047

2000 2001 2002 2003

1,450

2004

1,212

2005

1,765

1,046

2006

Capital outflows

Capital inflows

US capital inflows and outflows,* 19952006$ billion


24/112

2

Eiit 1.8

ASIA HAS BEEN THE LARGEST SOURCE OF CAPITAL INFLOWSTO THE UNITED STATES SINCE 2002

* After attributing funds channeled through the UK, Caribbean and Switzerland back to their original source.Bilateral capital inflows do not match total capital inflows due to statistical discrepancy.

** Includes Latin America, Africa, and unattributed flows.

Source: Bureau of Economic Analysis; US Treasury Department; McKinsey Global Institute analysis

Average20022005

$ billion

272

261

69

86

141

51

203

US bilateral capital inflows*$ billion

80 123166 124 152

177 110

121

120

203

210

279

7774

7

140

66

221

50

424

1998

165

2722

390

5

740

1999

207

1355

73

484

50

1,047

2000

151

13

70

306

47

783

2001

27

236

279

59

798

2002

7943

231

137

20

865

2003

262

287

406

43

1,450

1258

2004

52

290

264

1,212

2005

Rest of world**

Middle East

Other Asia

Japan

Europe

United Kingdom

Canada

Eiit 1.9

FOREIGN INVESTORS FROM EACH REGION BUY DIFFERENT TYPES OFUS ASSETS

Source: Bureau of Economic Analysis; US Treasury Department; McKinsey Global Institute analysis

Inflows to the US by region, 20022005 average%, $ billion

24

69

37

23

6

46

18

12

61

70

12

17

18

32

Rest of

world

57

3 5

Middle

East

33

7

2203261

Lending/deposits

Government debt

Asia

16

53

Japan

1015

0

100% = 27214151

Private debt

Equity

FDI

United

Kingdom

25

86

Europe

19

44

8

21

23

4

69

Canada


25/112

25

This picture o oreign investment in the United States does not, however, revealthe whole story about who is unding the US current account decit. Although

many European investors are buying US equities and bonds, their countries may,

in act, be running current account decits themselves. The United Kingdom,

or instance, has invested an average o $11 billion in the United States rom

200205 while at the same time running its own current account decit. This

meant that the United Kingdom was a net importer o capital rom the world

during this period.

cuties it uet ut suuses

The ultimate answer as to who is unding the US current account decit is

countries with current account surpluses. Even i these countries do not invest

directly in the United States, their surpluses add to global liquidity and their

capital outfows lter through global nancial markets, ultimately enabling capital

infows to countries running current account decits.

Western Europe, Asia, and the oil-exporting economies are the main sources o

current account surpluses in the world today. In 200 net capital outfows rom

these countriescapital outfows minus their capital infowsreached $1.3

trillion (Exhibit 1.10).5 Oil exporters including the Middle East, Norway, Russia,

Nigeria, and Venezuela had the largest share ($8 billion), ollowed by East Asia

($ billion) and Western Europe ($308 billion).

The worlds net suppliers o capital have shited over time. Petrodollars rom

oil-exporting nations are a new and growing provider o the capital to the world as

a result o the increased price o oil in recent years. Over the past 15 years it was

East Asia that was the largest provider o capital to the world. During the 1990s

the net capital outfows rom East Asia came largely rom Japan. However, since

the nancial crises o 199798 many other Asian countries have started to run

current account surpluses and are today major providers o capital.

Some countries in Western Europe have also been large net exporters o capital

over the last 15 years, including Germany, Switzerland, and the Netherlands.

Unlike Asia, where capital outfows are largely in the orm o oreign-reserve

assets owned by central banks, European capital outfows come mainly rom the

private sector and have been invested broadly across debt and equity classes.

Many other countries in Western Europe run current account decits, notably

Spain, Ireland, Greece, and Portugal. Within the eurozone some o the capital

5 This is equal to the size o their current account surpluses, except or errors and omissions,

debt orgiveness, and other capital transers.


26/112

2

outfows rom surplus countries have undoubtedly gone to other countries within

the monetary union.

At $8 billion, the US current account decit is now equal to nearly 70 percent o

the net capital outfows rom countries with current account surpluses. This very

high proportion is historically unprecedented over the last 35 years (Exhibit 1.11).

h eig it ifs eet te Uite Sttes

Running a current account decit allows the United States to consume more

than it otherwise could while maintaining a high investment rate. It also keeps

interest rates lower than they might have been in the absence o these capital

fows. Academic researchers have conrmed that oreign capital infows into the

United States have lowered interest rates.7

The downside o the US current account decit is that the United States is

running up a large oreign debt. As in any household, a decit allows a country

In MGIs report Mapping Global Capital Flows: Third Annual Update, we report that the USabsorbs 85 percent o the worlds net capital fows. The reason or this discrepancy is that,

in that analysis, we looked at the current account positions o regions rather than countries.

This produces a smaller gure or global net capital fows, since some regionsnotably

Europehave both countries with current account surpluses and countries with decits that

cancel each other out.7 For example, Francis Warnock in How might a disorderly resolution o global imbalances aect

global wealth? 200, or Francis Warnock and Veronica Cacdac Warnock, International capitalfows and interest rates, September 200.

Eiit 1.10

ASIA, EUROPE, AND THE OIL ECONOMIES ARE THE WORLD'SNET CAPITAL EXPORTERS

Note: Only includes countries in any given year with a current account surplus/capital account deficit.

* Includes Algeria, Indonesia, Iran, Nigeria, Norway, Kuwait, Libya, Russia, Saudi Arabia, Syria, United ArabEmirates, Venezuela and Yemen.

Source: International Monetary Fund; Global Insight; McKinsey Global Institute Global Capital Flows Database

Net capital outflows from countries with current account surpluses$ billion

133 98 156240

218

198

169 224

292

359

435

446

110 132151

121125

89

91129

184

273

268

308

926

1,319

371

301

691

351

1,199

2006E200520042003

503509417412

484

81

429

WesternEurope

Petro-dollars*

Rest ofworld

EastAsia

2002

69

238

56

167

49

8

35

4

1995

9

62

1996

2

42

1997

3282

1998

60

1999

30

192

7

30

127

2001

42

108

2000


27/112

27

to shit consumption rom the uture to the presentbut creates liabilities that

need to be paid later. Interestingly, however, the US net debt position has not

deteriorated as quickly as the sum o these cumulative current account decits

would normally indicate. Since 2002 the US net debt position has risen by only

$275 billion while the current account decits over the period totaled $2,88

billion. This is because US oreign assets have appreciated in value at a higher

rate than oreign liabilities due to a small decline in the dollar over the period,

higher returns on oreign than US stock markets, and very high returns earned

by US companies on their oreign investments abroad. Overall then, the United

States has ound a way to consume more than it otherwise couldand to do so

relatively cheaply.

Running a persistent current account decit also risks aecting adversely the

mix o jobs available in the economy. While running a decit, the United States

is exporting less and importing more than it otherwise would. Where jobs in

tradable sectors such as manuacturing, automotive, and high-tech are well-paid

or otherwise desirable, the risk is that the quality o these jobs erodes. This

may partly explain why, even at a time o ull employment, many US workers still

express concern about oshoring and trade.

Finally, running a current account decit requires the United States to attract

oreign capital infows each year. I capital infows were to experience a sudden

Eiit 1.11

THE UNITED STATES HAS ABSORBED ~70 PERCENT OF NET GLOBALCAPITAL FLOWS SINCE 2001

* Line shows value of US current account deficit divided by the sum of net capital outflows minus inflows forcountries with a current account surplus in that year.


0

20

40

60

80

100

1970 1975 1980 1985 1990 1995 2000 2005

US current account deficit as % of global net capital outflows, 19702005%


28/112

28

stop, or even i oreign investors had less appetite or dollar assets in comingyears, the United States could potentially see interest rates rise and economic

growth slow down.

wt s use te US uet ut eit t g?

Several explanations have been oered or the growth in the United States cur-

rent account decit, and each contributes a part o the story. One explanation is

that the United States spends too much and saves too little. A current account

decit represents the gap between what a country saves and what it invests.

While US private and government investment rates have been relatively stable

over the past decade, the national savings rate has decreased. US corporations

have maintained healthy savings rates, but the US household savings rate has

allen by 5 percent over the past 15 years as consumers have gained greater

access to all kinds o credit productscredit cards, car nancing, mortgages,

home equity loans, and student loans. In addition, the large government scal

decits o recent years have caused government saving to turn negative, urther

lowering the national savings rate. A low level o domestic savings has allowed

the United States to consume more than it otherwise would. By running a current

account decit, the United States has generated enormous demand, oering

other countries opportunities to export.

An alternative explanation, amously asserted by chairman o the US Federal

Reserve Benjamin Bernanke, is that a global savings glut has caused oreign

investments to fow into the United States.8 This has caused the dollar and other

asset prices to rise, allowing the United States to run a current account decit.

Despite the relatively low interest paid to oreigners, the United States oers an

attractive risk-return prole. It has robust GDP - and productivity-growth rates or

a mature economy; deep and liquid nancial markets with strong creditor and

shareholder rights; a stable macroeconomic environment; and a currency thathas de acto been the global reserve currency and has provided a sae store

o value. One estimate has capital infows to the United States in recent years

amounting roughly to 1015 percent o oreign savingsabout the allocation to

dollar assets that one might expect.9

A nal explanation is that the growth in the US current account decit is a natural

consequence o the integration o world nancial markets. This process has

started to sever the link between domestic savings and investment, allowing

8 Benjamin Bernanke makes this point in The Global Savings Glut and the US Current Account

Decit, 2005; and in Financial Regulation and the Invisible Hand, April 11, 2007.

9 This estimate appears in Richard Cooper, Living with Global Imbalances: A Contrarian View,

2005.


29/112

29

many more companies and households to invest abroad. The corollary o this isthat more countries than ever beore are running larger current account decits

and surpluses.10

A related actor is the impact o globalization o industries. As companies ex-

pand into oreign markets and restructure their operations internationally, they

trade goods and services with their oreign subsidiaries. Such transactions are

counted as imports and exports even though these subsidiaries are owned by

parent companies within the country. Theoretically the sum o these oreign-

aliate imports and exports could balance out. However, in the case o the

United States, whose companies are expanding abroad more quickly than oreign

companies are expanding in the United States, previous MGI research has ound

that trade with oreign aliates accounted or one-third o the US current account

decit in 200.11

These explanations are complementary, and all play a part in the rise in the US

current account decit to its present level.

10 Although there is still a strong correlation between domestic savings and investment or large

economies, it has eroded over the last ten years or smaller countries. See Martin Feldstein,Monetary policy in a changing international environment: The role o capital fows, 2005.

11 Farrell, Diana, Sacha Ghai, and Tim Shavers, A silver lining in the United States trade decit,

2005.


30/112

30

a se k t US imts

The United States imported $2.2 trillion o goods and services in 200. O

this, $32 billion was in services and $1,85 billion was in goods. The US

International Trade Commission tracks the source o these imports to the

United States. In 200, the United Kingdom, Germany, Japan, and Canada

were the top our sources o service imports (Exhibit 1.12). The story or

goods imports is somewhat dierent. Here, the largest sources o US goods

imported were Canada ($303 billion), China ($287 billion), Mexico ($197

billion), and Japan ($18 billion).

However, this picture o the source o US goods imports is somewhat skewed.

Countries oten import components o the goods that they export. Previous

MGI research has shown, or instance, that as much as 70 percent o the

value o Mexicos electronics exports comprises imported components and

other inputs rom elsewhere in world.12 Thus, to understand precisely how

much the United States imports rom each country, the value added o goods

should be attributed back to its original source.

12 Farrell, Diana et al., New Horizons: Multinational Company Investment in Developing Economies,McKinsey Global Institute, October 2003.

Eiit 1.12

THE UNITED KINGDOM AND GERMANY ARE THE LARGEST SOURCE OFSERVICE IMPORTS TO THE UNITED STATES

* Includes Latin America, Caribbean, Middle East, Africa and unattributed services imports.


US services imports$ billion, 2006

40

28

25

25

83

55

70

Japan

Other Europe

Germany

Other Asia

Rest of world*

Canada

Mexico 16

United Kingdom

342


31/112

31

MGI looked at the United States three largest trading partners in goodsCanada, China, and Mexicoto develop a better picture o bilateral imports

into the United States. We chose the top three because Canada is the largest

and thus the most important trading partner, and because we oten think

o China and Mexico as assembly locations with actories assembling

imported components or export. For these three countries, we determined

what percentage o their exports to the United States, on average, were due

to imported inputs. We then attributed the value o those imported inputs to

an estimate o their original source. See appendix A or more detail on the

methodology.

Ater making this adjustment, we see that while Canada remains the single

largest source o imports to the United States, Japan surpasses Mexico and

becomes nearly as large as China (Exhibit 1.13). Europes share o US imports

increases as well. Finally, we see that the United States itsel accounts or

nearly $100 billion o imports. This refects the trend or US companies

to set up oshore production acilities, particularly in Mexico and Canada.

Parent companies oten send components and other services to their oreign

subsidiaries, and then import nished goods back into the United States.

Eiit 1.13

MEXICO AND CHINA CONTRIBUTE LESS TO US IMPORTS WHENACCOUNTING FOR THE VALUE OF IMPORTED COMPONENTS

US goods imports input imports accounted for$ billion, 2005

US goods imports direct view$ billion, 2006

303

197

287

148

187

347

170

206

Canada

1,845

Mexico

Japan

China

Rest of world*

Oil-exporters**

Europe

Other Asia

* Includes Latin America, Caribbean, Africa, non-oil exporting Middle East and unattributed imports.

** Includes Algeria, Indonesia, Iraq, Iran, Nigeria, Norway, Kuwait, Libya, Oman, Russia, Saudi Arabia, Syria,United Arab Emirates, Venezuela and Yemen.


259

101

198

175

225

380

185

232

91

Canada

1,845

United States

Rest of world*

Petrol**

Europe

Other Asia

Japan

Mexico

China

Re-attribute


32/112

32

a se k t US it ifs

The majority o oreign capital infows into the United States today is invested

in debt securities. In 2005 oreign purchases o US bonds amounted to $822

billion. O this $31 billion went into corporate bonds while $50 billion went

into treasury bills and other government bonds. Foreign investors have their

largest debt holdings in US retail, manuacturing, and media companies (Exhibit

1.1). I we compare the size o these holdings with the entire debt issued by

S&P 500 companies, we nd that oreign holdings are disproportionately high

in media, chemicals, computers, and retail while they are disproportionately

low in automotive, oil, and telecom.

Equity investments into the United States, including purchases o equity shares

and FDI into companies, totaled $197 billion in 2005. Foreign purchases o

equity shares have been concentrated in nancial institutions, computers and

electronics, and health care. FDI has been most prominent in manuacturing

and, more recently, in nancial institutions (Exhibit 1.15). Equity investments

benet the United States in many ways. They have created 5 million jobs over

the past 20 years; they have provided $27 billion or research and development

and $112 billion in private-sector capital investment; they account or roughly

20 percent o US exports; and they account or percent o US private-sector

Eiit 1.14

THE MAJORITY OF FOREIGN CAPITAL INFLOWS TO THE UNITED STATESIS INVESTED IN DEBT SECURITIES

Note: Numbers do not sum due to rounding.

Source: International Monetary Fund; US Treasury; McKinsey Global Institute Global Capital Flows Database

Net investments in US assets by foreigners$ billion, 2005

Foreign private debt holdings by sector, 2005%, 100% = $1.7 trillion*

Foreign equity holdings by sector, 2005%, 100% = $2.0 trillion*

194

287

219

316

87

110

1,212

2005

FDI

Equity

Private debt

Other govtdebt

Treasurybills

Lending/deposits

26

5

7

10 11

12

29

Financialinstitutions

All others

Healthcare/pharma

Retail

Media

Oil

Computers &

electronics

23

6

7

78

14

14

21All others Retail

Manufacturing

MediaAutomotive

Utilities

Computers &electronics

Healthcare/pharma


33/112

33

labor compensation at an average compensation level 31 percent higher thanthe national average.13

13 Matthew J. Slaughter, Insourcing Jobs: Making the Global Economy Work or America, Organiza-

tion or International Investment, Washington, DC, October 200.

Eiit 1.15

FOREIGN DIRECT INVESTMENT IN THE UNITED STATES IS DIVERSIFIEDACROSS SECTORS

FDI inflows to the United States by sector$ billion

Other sectorsincludeagriculture,mining,utilities,construction,transportation,and real estate

Source: Bureau of Economic Analysis; International Monetary Fund; McKinsey Global Institute Global Capital Flows

Database; McKinsey Global Institute analysis

38

84107

5329 22

57

20

58

26

40

37

26

28

144

22

14

15 6

1

35

2

21

315

8

12

123

179

1997

413

2

1998

6

56

84

1999

58

26

2000

10

26

54

290

2001

11

10

2002

12228

2

-1

2003

2

51

9

Manufacturing

2004

123

Services

Financial institutions

Information

Trade

11

27

106

12

131

83

65

321

167

2005

12

Other


34/112

3


35/112

35

2. Could the world und a larger US

current account decit?

For many years now economists and policy makers have been saying that the

US current account decit is unsustainably largeand yet it has continued to

grow. This prompted MGI to analyze whether the US current account decit could

continue to increase urther over the next ve years. I current trends persisted,

how large would it become? Would there be enough global capital to und a largerUS decit? Would the implied US net oreign-debt position be sustainable?

We consider it plausible that the US current account decit could continue to

grow or another ve years. I current trends were to continue, we nd that the

US current account decit would reach $1. trillion or 9 percent o GDP in 2012.

Although this is very large, we nd that under most scenarios a decit o this

size would be quite possible. Current account surpluses in other countries could

conceivably grow large enough to und a US decit o this size, and the result-

ing level o US oreign debt would still be manageable. In this chapter we also

consider a variety o downside scenarios that could limit growth in the current

account surpluses in other countries necessary to und a larger US decit.

US dEIcIT woUld rEach $1.6 TrIllIon In 2012 UndEr cUrrEnT

TrEndS

I current trends in global savings and investment continued or another ve years,

and i there were no adjustments in exchange rates, the US current account

decit would reach $1. trillion by 2012, or 9.0 percent o GDP (see appendix

A or the methodology used to arrive at this projection) (Exhibit 2.1). Under thisscenario the United States would have a trade decit in goods and mineral uels


36/112

3

o $1,2 billion, a surplus in services trade o $138 bill ion, negative net transerpayments o $15 billion, and negative net oreign income o $122 billion.

To und this growing current account decit the US net oreign debt would triple,

reaching $8.1 trillion in 2012, up rom $2.7 trillion in 200 (Exhibit 2.2). This

implies much aster growth o US net oreign debt than we have seen in the past

and may overstate the true gure. As explained in chapter 1, over the last ve

years the US net oreign debt has risen much less than the cumulative current

account decits would imply. This is due to the depreciation o the dollar over

that period, which raises the value o US oreign assets, and to the appreciation

o US oreign assets in oreign direct investment and equity markets abroad.Going orward, our orecast assumes no change in the value o the dollar and

takes the average rate o appreciation o US oreign assets and liabilities over

the last 15 years.1

Even so, the implied level o US net oreign debt in 2012 percent o

GDPis not unprecedented. Several countries currently have net external debt

o a similar size relative to GDP including Mexico, Australia, and Ireland (Exhibit

2.3). Countries experiencing nancial crises over the last ten years usually had

higher levels o debt. Examples include Argentina at 7 percent o GDP in 2002,

1 For more on why the US oreign debt has not risen aster, see the IMFWorld Economic Outlook,

April 2007.

Eiit 2.1

% ofGDP -4.5 -4.8 -5.7 -6.3 -6.5 -6.7 -7.2 -7.7 -8.1 -8.6 -9.0

IF CURRENT TRENDS CONTINUED, THE US CURRENT ACCOUNT DEFICITWOULD REACH $1.6 TRILLION BY 2012

Source: Bureau of Economic Analysis; International Monetary Fund; McKinsey Global Institute Global Capital FlowsDatabase; McKinsey Global Institute analysis

-479 -544-662

-779

-122

-100

-108

-116

-125

-134

-145

122 122 138

-91

-911-836

1089685

72-65

-84-855

2006

-41

-931

2007

-1,305

Goodstrade

12

Servicetrade

-1,590

-1,452 Transfers

Income

2012

-1,462

37

2011

-1,349

28

2010

-2258 -5

-64

-473

2002

49-7

-69-527

2003

5111

-82

62

2004

-1,014

-1,178

2009

-1,048

-665

2008

-1,238

-86-792

2005

-1,129

Forecast$ billion (nominal)


37/112

37

Eiit 2.2

THE US NET FOREIGN DEBT WOULD REACH TO $8.1 TRILLION IN 2012,OR 46 PERCENT OF GDP

Source: Bureau of Economic Analysis; International Monetary Fund; McKinsey Global Institute Cross-Border InvestmentsDatabase; McKinsey Global Institute analysis

-24 -32-21 -28-20 -36 -40-21-23 -21 -46

Forecast

% ofGDP

-8,107

-6,961

-5,918

-4,974

-4,128

-2,729-2,546-2,449

-2,340-2,454

-3,384

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Eiit 2.3

LEVEL OF US EXTERNAL DEBT IN 2012 WOULD BE HIGHBUT NOT UNPRECEDENTED

Source: International Monetary Fund; McKinsey Global Institute Cross-Border Investments Database; McKinsey GlobalInstitute analysis

Net foreign debt, % of GDP%

4645

51545456

64

74

78

85

Brazil2002

Ireland2005

Argentina2002

Sweden1993

Mexico2005

Thailand1997

Australia2005

UnitedStates2012

Hungary2005

Malaysia1997


38/112

38

Sweden at 5 percent in 1993, and Thailand at 78 percent in 1997. The UnitedStates is unique, however, in that its oreign debt is denominated in its own cur-

rency, thereby eliminating currency risk. Moreover the size and productivity o the

US economy, combined with the role that the United States plays as a hub to the

world nancial system, mean that it is an attractive place or oreign investors.

All these actors suggest that the United States may well be able to maintain a

larger oreign debt than other countries have in the past.

Nor would the implied interest payments on an external debt o this size be large

relative to the US economy. The United States has consistently earned superior

returns on its oreign assets compared with what it pays out on its oreign li-

abilities. Even using a smaller spread on returns going orward than the 15-year

average would imply, US net oreign-income payments in 2012 would, at $122

billion, be less than 1 percent o GDP (Exhibit 2.).2 Projecting current trends

even urther, we nd that it would take until 2020 beore US net oreign-income

payments would reach even the relatively modest level o 3 percent o GDP .

Furthermore a larger US decit would not imply a dramatic increase either in the

share o gross world savings being invested in the United States or in the propor-

tion o US nancial assets owned by oreign investors. Under current trends

2 In this calculation we use the 200 spread o 0.9 percent, rather than the 1.3 percent average

over the last 15 years.

Eiit 2.4

US NET FOREIGN INTEREST PAYMENTS WOULD AMOUNTTO LESS THAN 1 PERCENT OF GDP IN 2012

0.1 -0.1 0.0 -0.2 -0.3 -0.4 -0.5 -0.7

Source: International Monetary Fund; McKinsey Global Institute Cross-Border Investments Database; McKinsey GlobalInstitute Global Capital Flows Database; McKinsey Global Institute analysis

Net interestpayments%

Income receipts and payments as a % of GDP

3.8 4.65.0 5.3 5.6 5.9

6.3 6.6

10

2012

-7.3

2011

-6.8

2010

-6.3

2009

-5.9

2008

-5.5

-10

-5.0

2006

-4.7

2005

-3.7

2007

Incomereceipts

Incomepayments


39/112

39

gross capital infows into the United States would rise to 23 percent o projectedgross savings in the rest o the world, up only slightly rom the share in 200

(Exhibit 2.5). Foreigners would own 1 percent o US equities and 2 percent

o US bonds in 2012, up rom 1 percent and 20 percent respectively in 2005

(Exhibit 2.). Overall, while an increasing level o debt and oreign ownership in

the United States is less than desirable, a growth in debt over the next ve years

remains possible.

From the perspective o oreign indebtedness, then, there is no reason to believe

that the US current account decit could not continue to grow over the next ve

years. However, the question remains as to whether it is plausible that other

countries would have suciently large current account surpluses to und a US

decit o $1. trillion.

Global nET capITal oUTlowS woUld Top $2.1 TrIllIon In oUr baSE

caSE

To run a current account decit the United States must borrow rom abroad.

As we explain in chapter 1, net capital outfows rom countries with current ac-

count surpluses und the US decit. Under reasonable assumptions these could

indeed grow enough over the next ve years to und a $1. trillion US current

account decit. In our base case global current account surpluses would grow

to $2.1 trillion in 2012 barring major changes in exchange rates and oreign

savings rates. East Asia would account or $799 billion o this, Western Europe

$23 billion, and oil exporters $387 billion (Exhibit 2.7).3

We arrive at this projection by developing a base case or each o the three

major capital-exporting regions and then considering downside scenarios or

each. In East Asia, we base our projection on the average annual growth rate

o current account surpluses in the region between 1998 (the year ollowingthe Asian nancial crisis) and 200. At this growth rate, however, several East

Asian countries would have very large current account surpluses relative to their

GDP . China, or instance, would have a surplus o 17.2 percent o GDP compared

with 7. percent in 200. Although some smaller economies such as Singapore

currently have much larger current account surpluses than this, it seems unlikely

that either China or oreign governments would nd a Chinese surplus o this

3 This includes Algeria, Iran, Saudi Arabia, Kuwait, Syria, Yemen, UAE, Oman, Qatar, Indonesia,

Nigeria, Norway, Russia, Venezuela. Using the growth rate since 1998 also produces a lower, more conservative projection than i

we were to take growth since 2000 or 2002. The same is true or Europe, or which we use the

same methodology.


40/112

0

Eiit 2.5

THE US SHARE OF REST OF WORLD GROSS SAVINGS WOULDINCREASE SLIGHTLY OVER THE NEXT FIVE YEARS

Source: Bureau of Economic Analysis; Global Insight; International Monetary Fund; McKinsey Global Institute GlobalCapital Flows Database; McKinsey Global Institute analysis

22.822.221.5

20.819.9

19.119.7

15.1

20.0

14.015.215.6

20.4

2000 2001 2002 2003 2004 2005 2006 2007E 2009E 2010E2008E 2012E2011E

Rest of worldgross savings$ trillion

US capital inflows as a share ofrest of world gross savings%

12.15.1 5.0 5.3 6.2 7.3 8.0 9.0 9.4 9.9 10.4 10.9 11.5

Forecast

Eiit 2.6

FOREIGN OWNERSHIP OF EQUITIES AND BONDS WOULD INCREASEONLY SLIGHTLY OVER THE NEXT FIVE YEARS

Source: McKinsey Global Institute Global Financial Stock Database; McKinsey Global Institute Cross-Border InvestmentsDatabase

92 88 86 8074

100%

9

7.6

1990

12

10.8

1995

14

16.0

2000

20

23.8

2005

26

41.6

2012

Domestic-ownedbonds

Foreign-ownedbonds

Projected foreign ownership of US bonds, 2012%, $ trillion

92 92 89 86 84

100%

8

3.1

1990

8

6.9

1995

11

15.1

2000

14

17.0

2005

16

32.1

2012

Domestic-ownedequities

Foreign-ownedequities

Projected foreign ownership of US equities, 2012

%, $ trillion


41/112

1

size desirable. To make our projection plausible we thereore limit the amount bywhich a countrys current account surplus can grow.5 For most countries the cap

is 10 percent o GDP , which is exceeded today only by small economies. For very

small economies like Singapore and Hong Kong we set the cap at 27 percent o

GDP , the level in Singapore today. We use the same methodology or Western

Europe.

We project net capital outfows rom oil-exporting economies based on the uture

price o oil. In our base case their capital outfows amount to $387 billion in

2012slightly lower than the outfows o $8 billion seen in 200. The reason

or this decline is that we assume an oil price o $50 per barrel compared with

nearly $0 per barrel in 200. These assumptions are those used in MGIs

proprietary model o global energy demand. We translate oil revenues into net

capital outfows rom these economies based on a methodology developed by

Brad Setser (see appendix A or details).7

5 We assume smooth growth between now and 2012 or countries whose surpluses reach 10

percent o GDP. For nancial hubs, such as Singapore and Hong Kong, we cap current account

surpluses to 27 percent o GDP, the level o Singapores current account surplus in 200.

Curbing Global Energy Demand Growth: The Energy Productivity Opportunity, McKinsey Global

Institute, May 2007.

7 Brad Setser, Oil and Global Adjustment, 2007.

Eiit 2.7

WITH CURRENT TRENDS CONTINUING, GLOBAL CURRENT ACCOUNTSURPLUSES WOULD REACH $2.1 TRILLION IN 2012

* Includes Algeria, Indonesia, Iran, Nigeria, Norway, Kuwait, Libya, Russia, Saudi Arabia, Syria, United ArabEmirates, Venezuela and Yemen.

Source: International Monetary Fund; McKinsey Global Institute Global Capital Flows Database; McKinsey GlobalInstitute analysis

346388

437491

553623

121148

180

219

267

308

99

440

488

1,373

2007E

432

536

1,477

2008E

426

590

1,600

2009E

406

651

1,728

2010E

393

721

1,886

81

387

799

2,077

2012E

Rest of world

Western Europe

2011E

East Asia

484

446

1,319

2006E

Petrodollars*

Projection of global current account surpluses$ billion


42/112

2

Countries in the rest o the world also run current account surpluses that contributeto global net capital outfows. These include Canada, Brazil, Israel, Argentina, and

Chile, to name a ew. We project their current account surpluses using the same

methodology as that used or East Asia and Western Europe. In 2012 their current

account surpluses would reach $27 billion, up rom $81 billion in 200.

US SharE o Global capITal lowS woUld IncrEaSE

Even i global net capital outfows were to grow to $2.1 trillion in 2012, the US

current account decit would absorb a historically unprecedentedand perhaps

implausibleshare o this. With a decit o $1. trillion, the United States wouldtake up 77 percent o the worlds net capital outfows. In chapter 1 we saw that,

since 2001, the US decit has absorbed around 70 percent o the net capital

fows rom countries with current account surplusesitsel a new high (Exhibit

1.12). In the late 1980s the United States took up around 0 percent o global

net capital outfows or a ew years, but ater this the share ell. I the United

States continued to run very large current account decits or ve more years, it

would be the rst time in modern history that a single country continued to absorb

such a large share o the worlds capital outfows over such a prolonged period.

Is this scenario plausible? On one hand, larger current account decits in other

countries could limit the ability o the United States to und its decit. Australia

has had a current account decit or more than 25 years now, and the United

Kingdom also runs a sizeable decit. Some economies in Europe such as Spain,

Portugal, and Greece run decits too, unded in large part by those o their

eurozone neighbors with surpluses. On the other hand, our orecast o global

net capital outfows is conservative since we constrain the size o uture current

account surpluses relative to GDP . Moreover, the US decit has already broken

all historical precedents since 2001. With that in mind, a urther rise in the share

o world capital that the decit absorbs could be possible.

downSIdE ScEnarIoS or Global nET capITal oUTlowS

There are several downside scenarios that could limit global net capital outfows

in the years to come. We consider our o these scenarios in an attempt to gauge

the ability o the United States to und a decit o $1. trillion and to identiy

where the most important sensitivities lie.

1. ci ses ess; J ges

Today Asia, and particularly China and Japan, plays a key role in supplying the

worldand the United Stateswith capital. There are several scenarios under


43/112

3

which the current account surpluses o China and Japan could be reduced overthe next ve years, however, thus lowering their net capital outfows.

Many economic orecasters are predicting that Chinas current account surplus

will decline in years to come as domestic consumption rises and its savings

rate declines. Ater all Chinas current account surplus is a relatively new

phenomenon, growing rom just 1.3 percent o GDP in 2001 to 7. percent in

2005. Policy makers both in China and abroad have argued recently that Chinas

economy would benet rom a rebalancing away rom investment and exports

and towards a higher rate o domestic consumption. The widespread assumption

has been that this rebalancing will occur at least partly through a all in Chinas

historically high savings rate. Alternatively, China could ollow the advice o many

economists and global policy makers and allow the yuan to appreciate more

rapidly over the coming years. In either case Chinas current account surpluses

would grow more slowly.

Japans current account surpluses could decline i its aging households save

less. Japans household saving rate is already declining and, as more Japanese

retire, it will likely all urther. The combined eect o a lower level o savings in

both China and Japan would reduce East Asian net capital outfows. Instead o

reaching $799 billion in 2012, as in our base case, net capital outfows rom East

Asia would total only $27 billion (Exhibit 2.8).8 This would mean that the world

supply o capital would be only $150 billion larger than the US current account

decit in 2012, pushing US uptake o world capital to 91 percent. This level is

implausible given the preerence o oreign investors to vary their portolios in

terms o both geography and currencies.

We consider this outcome rather unlikely, however. MGIs research on Chinas

evolving consumer market nds that, although Chinas domestic consumption

will indeed rise strongly in coming years, the rise will derive largely rom rising in-

comes rather than rom alling savings.9 Indeed MGI projects only a slight decline

in Chinas national savings rate. Although Japans population is clearly aging, the

impact on savings rates over the next ve years will be minimal.

2. Suuses i Eue g me s

There are two scenarios under which net capital outfows rom Europe might be

smaller than projected over the next ve years. First, what i current account

8 We arrive at this gure by using Global Insights orecast or current account surpluses in Chinaand Japan, rather than our projection o continued growth.

9 From Made in China to Sold in China: The Rise o the Chinese Urban Consumer,McKinsey

Global Institute, November 200.


44/112

surpluses within Europe were to grow more slowly in the uture? Ater all, thesesurpluses are already becoming large relative to GDP in many o the main

economies. For instance, in 200 the Netherlands, Sweden, and Germany had

surpluses o 7.5 percent, 7. percent, and .7 percent o GDP respectively. I we

assume that uture current account surpluses in Western Europe grow in line with

projected GDP growth (2.1 percent annually)rather than at the growth rate o

their current account surpluses (15.1 percent annually) as in our base casenet

capital outfows rom Europe would be reduced by $270 billion in 2012.

An alternative downside scenario or Europe takes account o capital fows be-

tween eurozone countries. Given the common currency, it is likely that countries

with current account surpluses such as Germany are sending much o their or-

eign investment to countries with decits, such as Spain, Portugal, and Italy. One

study has estimated that around hal o the increase in eurozone current account

surpluses and decits over the last eight years was due to capital fows within

the eurozone.10 I we assume that 50 percent o eurozone surpluses remain

within the region and we exclude these rom our sum o global net capital fows,

Western Europes net capital outfows in 2012 would be reduced by $223 bil-

lionrom $23 billion to $00 billion. Thus Western Europes capital outfows

10 Alan Ahearne et al., Internal and External Current Account Balances in the Eurozone, February

2007.

Eiit 2.8

UNDER DOWNSIDE SCENARIO, EAST ASIAN CURRENT ACCOUNTSURPLUS WOULD SHOW NO GROWTH THROUGH 2012

Source: Global Insight; McKinsey Global Institute Global Capital Flows Database; McKinsey Global Institute analysis

200

300

400

500

600

700

800

2 00 6 2 00 7E 2 00 8E 2 00 9E 2 01 0E 2 01 1E 2 01 2E

Base caseCapital projected supplygrowing in line with 19982006compound annual growth rateof current account surplus

Aging JapanReduced capital supply fromJapan; all other countries

according to base case

Adjusting ChinaReduced capital supply fromChina; all other countriesaccording to base case

Scenarios

372

799

427

Projection of East Asian current account balances$ billion


45/112

5

over the next ve years could be reduced by $220 billion to $270 billion lessthan we project in our base case. This would raise the US absorption o world

capital to a very high 88 percent.

3. ee ets ue t ie i

Another possibility causing concern among commentators and economists is

reduced net capital outfows rom oil-exporting economies in uture years as a

result o lower oil prices. However, we nd that changes in the price o oil would

not signicantly aect the extent to which there is sucient capital to und a US

current account decit. This is because o petrodollar recycling. I the price o

oil increases the US trade decit rises, but net capital outfows rom oil-exporting

countries rise too. At the same time, current account surpluses in Europe and

Asia all somewhat due to more costly oil impor ts. Conversely, i the oil price alls

the decline in capital outfows rom oil-exporters is oset by a reduction in the

US current account decit and by current account surpluses in Asia and Europe

rising on the back o cheaper oil imports. We nd that the US share o net capital

outfows varies by less than 5 percentage points between conditions under which

oil is priced at $30 per barrel and at $70 per barrel (Exhibit 2.9).

Even so, a higher price o oil does make it slightly more dicult or the United

States to und a decit. Under the downside scenario we posit oil prices rising

to $70 per barrel instead o $50 per barrel as in our base case. I this were to

Eiit 2.9

860 9671,107 1,263

1,416 1,5801,720

2,1551,9711,8071,6611,5171,3961,320

0500

1,0001,5002,0002,500

VARYING THE OIL PRICE HAS MINIMAL IMPACT ON THE RATIOOF THE US CURRENT ACCOUNT TO WORLD CAPITAL

US CA deficit

World capital

US current account deficit vs. world supply of capital$ billion

69

$70 per barrel scenario

US % of total 73 76 78 80 8065

2007 2008 2009 2010 2011 20122006

1,5911,4521,3051,1781,047931864

1,320 1,373 1,4771,600 1,728

1,886 2,077

0500

1,0001,500

2,0002,500

68

Base case $50 per barrel

US % of total 71 74 76 77 7765

860 891 9801,107 1,178

1,306 1,443

2,0171,8181,6631,5631,4081,3551,320

0500

1,0001,500

2,0002,500

$30 per barrel scenario

US % of total 66 70 71 71 72 7265

Source: McKinsey Global Institute GEM model; McKinsey Global Institute Global Capital-Flows Database; McKinseyGlobal Institute analysis


46/112

happen, net capital outfows rom oil-exporting countries would increase by $21billion, but the US decit would also grow by an additional $129 billion. Current

account surpluses in Western Europe and East Asia would be reduced by $1

billion and $102 billion respectively. The overall eect is that global net capital

fows would increase by $78 billion and the US decit would rise by $129 billion.

This would barely aect the ability o the United States to und a larger current

account decit.

4. rest s sme suuses

Our nal downside scenario considers other countries in the world that are run-

ning current account surpluses such as Canada, Brazil, and Argentina. Construct-

ing a rational downside scenario or the rest o the world is dicult since such

a group includes a large and diverse set o countries (15 countries in 2005).

Even i one or more o them experienced slower growth in their current account

surpluses over the next ve years, other countries might see their surpluses grow

aster than in recent years. Moreover, some countries current account decits

might turn into surpluses. Indeed, we nd that surpluses in our rest o the

world category have grown steadily since 1998 despite changes in individual

countries in that group.

Despite recent historical experience, we still decided to test slower growth in the

current account surpluses o this group in order to understand the sensitivity to

this o the US decit. In our base case we assumed that their current account

surpluses grow at the same rate as they have since 1998. However, i we project

a slower growth rate usin

mgi us imbalancing act full report

Documents