The Hidden Key to the Growth

48 THE INTERNATIONAL ECONOMY WINTER 2006
The Hidden
Key to
Growth
How local services stimulate economic expansion.
Having focused for many years on manufacturing-led
growth, policymakers across the developing world now
recognize the contribution that service exports can make.
India leads the world in offshore IT services. Dubai has
tourism as well as a growing financial services hub.
Singapore is building hospitals to serve patients from
across Asia. The Philippines is developing call centers.
Yet these offshore service strategies overlook a far larger,
if less well-understood, opportunity to boost wealth creation: stimulating domestic
service sectors.
Higher productivity in services is the key to growth in any economy. Local services
account for more than 60 percent of all jobs in middle income and developed
economies, and virtually all of new job creation (Figure 1). Manufacturing is not
going to be a sustainable long-term source of new jobs anywhere—even in China—
given the rapid advances in technology and productivity that are reducing industry’s
labor needs.
Why, then, do so many policymakers omit local services from their development
plans? Part of the reason is that service work has a poor reputation. Low-skill,
low-wage, ephemeral jobs in fast food joints and beauty parlors hardly seem the
By Martin Baily, Diana Farrell, and Jaana Remes are with McKinsey Global
Institute.
BY MARTIN BAILY, DIANA FARRELL, AND JAANA REMES
THE MAGAZINE OF
INTERNATIONAL ECONOMIC POLICY
888 16th Street, N.W.
Suite 740
Washington, D.C. 20006
Phone: 202-861-0791
Fax: 202-861-0790
www.international-economy.com
WINTER 2006 THE INTERNATIONAL ECONOMY 49
BAILY, FARRELL, AND REMES
building blocks of a modern economy. But such jobs
form the minority of service employment: even in
the United States, widely thought to have too many of
them, they represent only 22 percent of the huge
range of total service employment. In fact, services
comprise many activities critical to economic growth,
like power supply, transport, and telecommunications,
as well as numerous high-skill, high-wage
occupations, such as accountants, researchers, and
professionals in health and financial services.
After years of neglect and undue regulatory constraints,
local service productivity in most emerging
economies lags far behind productivity in sectors
developed for export. This is a pity. Research by the
McKinsey Global Institute suggests that, given the
right competitive environment, local services across
the range can be a powerful source of wealth creation
and jobs for middle-income economies, more powerful
than offshore services could ever be.
FASTER GROWTH AND MORE GOOD JOBS
Once an economy reaches the middle income level of
development, service industries become a more
important source of job growth than manufacturing.
And, contrary to popular belief, a substantial percentage
of these jobs are high-skill and high-wage.
The more dynamic and competitive an economy’s
service sector, the more jobs and GDP growth it will
create.
More good jobs. Since 1997, employment has
declined in the goods-producing sectors of most
developed and many developing economies, leaving
service industries responsible for all net job cre-
Market service activities in
today’s economies are
tremendously diverse.
The EU Services Directive divides
them into three categories: services
provided to consumers, services
provided to other businesses, and
services provided to both consumers
and businesses.
In the United States, which is
typical of a developed service economy,
some 29 percent of the
roughly 100 million service jobs
are in consumer services, which
include retail, food, and accommodation
services, and personal services
like car repair shops, dry
cleaners, and beauticians. About 77
percent of U.S. consumer service
jobs are in relatively low-skilled
sales and service occupations.
These jobs tend also to have a
higher share of very small businesses,
higher business turn-over
rates, and a disproportional share of
female employees.
As economies grow richer,
business-to-business services represent
an increasing share of total
economic activity. Today, they represent
27 percent of all U.S. service
sector employment, almost as
much as consumer services. These
activities include: professional services,
such as law, accountancy,
and consulting; technical services
such as IT and software support;
wholesale trade services; and
employment services like headhunters
and temp agencies. The
recent rapid growth in business services
in developed economies is an
outcome of specialization. As companies
focus increasingly on their
core competencies, they buy more
non-core services from third parties.
Services provided to both consumers
and businesses include real
estate and banking, as well as services
based on extensive physical
networks, like telecommunications
and electricity supply. These types
of services account for another 7
percent of service sector jobs. The
remaining 36 percent of service
jobs are in non-market activities
like healthcare, education, and public
sector services.
What Are Services?
Even China, the world’s “factory
floor,” lost 15 million
manufacturing jobs, equivalent
to 15 percent of total
Chinese manufacturing.
50 THE INTERNATIONAL ECONOMY WINTER 2006
BAILY, FARRELL, AND REMES
ation. Among middle- and high-income economies
today, services generate 62 percent of all employment
on average, and the higher a country’s GDP
per capita, the higher the share of service employment
(Figure 2).
Manufacturing employment is shrinking worldwide
as a result of more efficient use of labor,
automation, and new IT. Roughly 22 million manufacturing
jobs disappeared worldwide between 1995
and 2002, despite policy efforts to preserve them.
Even China, the world’s “factory floor,” lost 15 million
manufacturing jobs, equivalent to 15 percent of
total Chinese manufacturing and a higher proportion
than the global average loss of 11 percent. New jobs
created by the boom in foreign manufacturing investment
were not enough to offset these losses, caused
largely by restructuring in China’s state-owned manufacturing
plants.
Somewhat surprisingly, service industries actually
create more high-skilled occupations than manufacturing.
In the United States, more than 30 percent
of service jobs are in the highest skill category of
professional, technical, managerial, and administrative
occupations. In contrast, only 12 percent of all
manufacturing jobs are in this category, and the same
pattern holds in other developed nations. There are
also many well-paid “blue-collar” jobs in services,
such as electricians, plumbers, and auto mechanics.
In fact, the distribution of wages in the United States
looks broadly similar in services and manufacturing.
There are more low-wage jobs in services, but also
many high-wage jobs, and the variance within each
sector is actually greater than the variance between
them. Moreover, the experience of some countries in
Europe shows that trying to contain growth in lowskill
service jobs by imposing high minimum wages
and other labor market restrictions results in higher
overall unemployment, not more high-skill jobs.
Low-skill consumer service jobs, just like lowskill
manufacturing jobs, may not be the most attrac-
1 Service Sector Growth During Economic Evolution
Percent of GDP, 1970–2001
Services
Industry
Agriculture
0
10
20
30
40
50
60
70
80
2000 1994 1988 1982 1976 1970
0
10
20
30
40
50
60
70
80
2000 1994 1988 1982 1976 1970
0
10
20
30
40
50
60
70
80
2000 1994 1988 1982 1976 1970
Low-Income Countries Middle-Income Countries High-Income Countries
*Industry includes manufacturing, mining, and construction; services include personal, professional, and public sector services and utilities.
**The World Bank defines middle-income economies as those with per capita GNI in 2003 between US$766 and US$9,385 measured with average
exchange rate over past two years.
Source: World Bank; World Development Indicators.
WINTER 2006 THE INTERNATIONAL ECONOMY 51
BAILY, FARRELL, AND REMES
tive. But they are crucial to all
economies in providing formal
employment for new entrants
to the workforce and also
unskilled workers—a group
whose only alternatives are
informal (and therefore illegal)
work or welfare. Even if consumer
service workers learn
few value-adding skills “on the
job,” having a formal position
can help them or their dependents
to study elsewhere, and
so move up the occupational
ladder.
Faster growth. Because of
their sheer size, local service
sectors like retail and construction
are important drivers of
overall GDP growth. And
access to high-quality local services
affects rates of growth in
all other sectors because every
enterprise uses them. Good
local services can also make a
difference in attracting foreign
direct investment. Electricity,
communications, and transport
quality and cost all influence
the overall attractiveness of an
offshore location to multinational
companies choosing where to invest. In the
early 1990s, for instance, India’s nascent offshore
sector was hobbled by unreliable phone and Internet
connections. It was only after local telecom services
improved that offshoring in India took off. Indeed,
McKinsey Global Institute interviews with executives
at multinationals show that they value a stronger
infrastructure and reliable network services more than
direct incentives from governments. Managers at
Brazilian auto OEMs and Indian offshore service
companies that received direct government incentives
told us they would rather the money had gone
on improving ports and roads in Brazil and telecommunications
in India.
MYTHS ABOUT SERVICES
Given the employment and growth benefits of a
dynamic service sector, why have so few economies
adopted service sector reforms? Three myths explain
this anomaly.
Myth 1. There is little scope for innovation in local
services, so reforming them won’t do much for overall
growth.
History shows otherwise. Productivity improvements
in service industries like electricity supply and
telecommunications were important drivers of overall
productivity growth in the developed economies
after World War II. In the United States, the late
1990s boom in productivity was in large part due to
services industries like retail, wholesale, and financial
services.
Indeed, McKinsey Global Institute’s studies of
countries around the world show that gaps between
productivity levels in their large, employmentintensive
local service sectors, such as retail and construction,
explain a substantial amount of the gaps
between their respective GDP per head figures. In
Korea, McKinsey Global Institute found retail sector
labor productivity was only 32 percent of the U.S.
2 Share Of Services Is High and Increases With GDP Per Capita
0 5000 10000 15000 20000 25000 30000 35000 40000
0
10
20
30
40
50
60
70
80
90
100
Cameroon India
Bangladesh China
Pakistan Romania
Philippines
Kenya
Peru
Malaysia
Egypt
Morocco
Venezuela
Columbia
Portugal Poland
Mexico
Chile
Brazil
Slovakia
Czech Rep.
Taiwan
New Zealand
Spain
Saudi Arabia
Israel
Netherlands
Hong Kong
Greece Korea
Russia
Italy Germany Japan
Finland
France
Denmark
Switzerland
Canada
Ireland
Norway
U.S.
Turkey
Thailand
Hungary
Indonesia
Share of services, GDP*
GDP per Capita at PPP, US$
Ukraine
*Includes all other sectors except agriculture, manufacturing, and mining.
Source: World Development Indicators, Global Insight.
52 THE INTERNATIONAL ECONOMY WINTER 2006
BAILY, FARRELL, AND REMES
level, compared with an average of 53 percent across
all the sectors we studied there. In Turkey, labor productivity
in manufacturing averaged 64 percent of the
U.S. level, while it was only 33 percent in services.
Retail sector reforms are particularly important
in triggering productivity growth, partly because
these sectors employ so many people, partly because
improvements here stimulate productivity advances
among upstream suppliers. For example, the liberalized
retail sector in the United States has been one
of the top three contributors to aggregate productivity
increases since 1995. Research has shown that
removing restrictions on outlet size, opening hours, or
product selection from retailers in other OECD countries
would allow their retailers likewise to streamline
distribution systems and grow both sales volumes
and employment. Their consumers, too, would benefit
from lower prices and a broader array of services.
Research elsewhere has demonstrated that liberalizing
trade policies governing services generally has
far higher welfare benefits for developing economies
than equivalent reforms to manufacturing or agricultural
policies. Even though trade barriers in services
are usually lower, the economic benefits from removing
them are larger because of the huge improvements
in service productivity they unleash.
Myth 2. Increasing service sector productivity will
rapidly increase unemployment.
This anxiety centers on the retail sector, a huge
employer in all economies. Policymakers rightly
believe that more productive supermarket and large
discount formats will drive out traditional, less productive,
small stores. But this is the normal process of
economic development that will result in a bigger
national income and higher overall employment.
McKinsey Global Institute emerging country
case studies show that in most cases net employment
in retailing increases when the sector adopts more
productive formats. Supermarkets and large-scale
retailers, because of their higher productivity, can cut
prices, attract more customers, and so increase their
incomes. As they grow, they employ more people.
Their growth also stimulates new jobs in retail supply
industries, such as food processing and consumer
manufacturing. In Mexico, for example, rapidly
expanding formal convenience stores were the main
source of employment growth in the retail sector after
it was opened to foreign investment. Likewise, in
Thailand and Poland the net impact on employment
of opening the retail sector to investment by modern
format retailers was likely to be neutral or positive.
This phenomenon helps explain why the share of
retail in total employment is still higher in the United
States, with its very small percentage of traditional
retailers, than in most low- and middle-income
economies, with a large proportion of retail workers
employed in traditional formats.
Myth 3. Services are an unreliable source of jobs.
Many policymakers still believe that manufacturing
jobs are not only higher skill and higher wage
than service jobs but also more reliable, because the
fixed costs of capital-intensive plants means they are
unlikely to move elsewhere. Are they?
There certainly is higher turnover in service jobs
than in manufacturing jobs. But service jobs provide
a much more reliable source of overall employment
than manufacturing. In any given year, on average
roughly 10 percent of all jobs in an economy come
to an end, because workers quit or become redundant.
More jobs end in services than manufacturing, particularly
in service segments dominated by small scale
operations, with their relatively high failure rates.
However, service industries as a whole create more
jobs than they lose, often through the activity of new
The experience of some countries
in Europe shows that trying to
contain growth in low-skill service
jobs by imposing high minimum
wages and other labor market
restrictions results in
higher overall unemployment,
not more high-skill jobs.
WINTER 2006 THE INTERNATIONAL ECONOMY 53
BAILY, FARRELL, AND REMES
entrants. Creating a dynamic service sector therefore
reliably guarantees lifetime employment opportunities
for everyone, if not the same job for life.
For example, from 1977–87, the U.S. auto repair
industry lost 49 percent of its jobs, but at the same
time took on new employees in jobs equivalent to 56
percent of total employment in the industry. So
although almost half of all auto repair jobs ended
over the period, net employment in the sector grew
by 7 percent. Data from middle-income economies,
albeit limited, suggest their dynamics of service job
destruction and creation are similar.
These myths about services have led many policymakers
to penalize or neglect their local
service sector, usually in favor of industrial
expansion. But McKinsey Global Institute’s country
research has shown such choices bear considerable
costs. Take Japan. By the end of the last century,
Japan’s world-class manufacturers of autos, steel,
machine tools and consumer electronics were legendary
for their performance. But their output comprised
only 10 percent of GDP. Productivity in the
rest of the economy—about 68 percent of it in local
services—was a dismal 63 percent of U.S. levels.
Low productivity in local services goes a long way
towards explaining why Japanese GDP growth tailed
off in the 1990s, just as subsequent incremental
reforms of service sectors help to explain the recent
improvement in Japan’s economic performance.
Japan’s Cabinet Office calculates that deregulation
in telecoms, transport, energy, finance, and retailing
were responsible for 4.6 percent of the country’s GDP
in 2002.
Neglecting and over-regulating service industries
also has the perverse consequence of encouraging
growth in the informal service sector. In many developing
and middle-income economies, the majority of
output in services such as retail and construction
comes from informal firms that neither pay full taxes,
nor abide by worker safety and other regulations, nor
even register. Their unearned cost advantage allows
low-productivity players to survive and prevents more
productive formal companies from taking market
share. In Brazil’s food retail sector, for example,
McKinsey Global Institute found that tax-paying, productive,
modern format retailers are at a large cost
disadvantage to small, unproductive, informal players
that evade tax. Acquiring the informal players is no
remedy as the scale gains would be too small to offset
the extra tax burden. As a result, growth in more
productive, higher-wage jobs in services is capped,
and average productivity in Brazil’s retail sector is
only 16 percent of the U.S. level.
HOW TO DEVELOP A DYNAMIC
LOCAL SERVICE SECTOR
Government policymakers keen to unlock local services’
power to generate growth and jobs must
remove barriers to competition. This requires leveling
the playing field so that services can compete
freely for capital, labor, and technology; removing
inappropriate restrictions on service businesses; and
tackling informality.
Level the playing field. Governments first need to
remove any remaining biases against services and
give them equal treatment in fiscal, financial, and
development policies. Then service companies can
compete for capital and workers on the same terms as
manufacturing firms.
Open up capital markets to local services.
Investments in services should be assessed against the
same criteria as manufacturing investments. In many
countries, this is not the case. Directed lending policies
in banking remain prevalent. In South Korea, for
instance, during its push for manufacturing-led
growth, banks were prohibited outright from lending
to consumer service sectors like leisure and real estate.
In China today, state-owned manufacturing enterprises
account for 65 percent of all loans, even though
they produce just 35 percent of industrial output.
Meanwhile, IPO rules in most developing countries’
stock markets allow only large companies to list,
thereby discriminating against service sectors that
have smaller players. Policymakers need to liberalize
financial systems so they will allocate capital to the
projects with highest returns, regardless of sector.
A dynamic service sector reliably
guarantees lifetime employment
opportunities for everyone,
if not the same job for life.
54 THE INTERNATIONAL ECONOMY WINTER 2006
BAILY, FARRELL, AND REMES
End industrial subsidies. Governments often explicitly
subsidize investments in manufacturing activities
they consider in the national interest. For instance,
Malaysia has supported the creation of the Proton in
order to have a national car company. The Brazilian
government offered subsidies worth $100,000 per job
created to foreign carmakers to invest in local factories,
prompting so much investment, the result was overcapacity
and generally low productivity throughout the
industry. Such subsidies not only often waste taxpayers’
money, but put service industries at a disadvantage.
Expand favorable business conditions to cover services.
Many middle-income economies have created
special economic zones for foreign and/or export manufacturers,
with more favorable tax and tariff rates and
lighter regulation than domestic companies face. It
makes sense to provide conditions that allow businesses
to flourish, but why not extend them to all businesses?
Governments should equalize regulations between sectors
and set corporate taxes at affordable levels across
the economy.
Physically integrate manufacturing and service activities.
Today’s special economic zones are often geographically
as well as fiscally separate from local
service providers, making them harder to serve. This is
one challenge facing service providers in Mexico’s
business centers, far away from the maquiladoras on
the U.S. border. Extending special economic zone-type
conditions to all businesses has the added advantage of
allowing them physically to reunite. This will be
increasingly important as manufacturers continue to
outsource more functions to third-party service
providers.
Remove the product market barriers limiting competition
in services. McKinsey Global Institute productivity
studies have shown that inappropriate product
market regulations governing service sectors are the
biggest barrier to increased competition, which drives
the diffusion of more productive processes. Product
market regulations govern company ownership, trade,
foreign direct investment, land use, prices, and products.
Misconceived regulations make competition less
intense by limiting the entry of new players (particularly
global ones), discouraging innovation among
existing competitors, and restricting enterprise scale.
Reduce public-sector ownership. Utilities, telecommunications,
and banking remain in government hands
in many emerging economies. Lack of investment and
low productivity in these sectors stunt not only their
own but also their customers’ growth. According to
some estimates, Mexico has forgone $50 billion of
potential investment in the electricity grid because this
has been entirely state-controlled since 1933.
Remove barriers to FDI in services. This can open the
door to substantial inflows of capital. Moreover, it
allows countries to benefit from the best practices and
increased competition provided by global companies,
which will impel service productivity upwards. For
example, when foreign direct investment restrictions in
retail banking were removed in many Latin American
countries during the 1990s, foreign companies invested
over $50 billion in their banking sectors alone over the
next ten years. Similarly, in India, foreign investment
played a key role in introducing competition to the formerly
protected auto assembly sector after it was liberalized
in 1991 and helped to spur productivity
improvements.
Revise unnecessary barriers to scale. Scale can yield
substantial productivity gains to enterprises. Yet many
companies face limits to scale, like restrictions on store
size and land use, which keep them less productive than
they could be, without always yielding a commensurate
social gain. Productivity in housing construction,
for example, depends critically on scale. Yet in
Germany and France, construction companies cannot
acquire lots of land big enough to support large-scale
housing developments. This explains why productivity
in the German and French construction industries
lags far behind its equivalent in both the Netherlands
and the United States. Land purchase is similarly difficult
in many emerging economy cities because these
are large and crowded, and land titles are unclear. The
solution here is to clarify titles, so land can be traded
French zoning laws have required
retailers to get local authorization
before opening new stores bigger than
three hundred square meters.
WINTER 2006 THE INTERNATIONAL ECONOMY 55
BAILY, FARRELL, AND REMES
more easily and put to its most productive use. Many
governments also restrict store sizes to protect momand-
pop stores from large-scale retail outlets, but at the
cost of higher retail productivity. For instance, French
zoning laws have required retailers to get local authorization
before opening new stores bigger than three
hundred square meters, or expanding existing stores.
Enforcement of fiscal and administrative rules to
reduce informality. The high proportion of small firms
in service industries makes them particularly likely to
operate informally, ignoring tax requirements,
employee benefits, and other regulations. This is a much
larger barrier to growth than most policymakers in
emerging—and developed—economies acknowledge.
Steps to reduce informality in local service sectors will
be rewarded by rapid increases in their productivity,
growth, and employment.
Strengthen enforcement. Most informal business
evade taxes and bend rules because they can get away
with it. Strengthening inspection and audit services as
well as increasing penalties for rule-breaking will help
push enterprises into the formal sector.
Eliminate red tape. So will streamlining what businesses
must do to comply. For example, lots of companies
never register because the process is so long and
complicated. The noted economist and author Hernando
de Soto found that in Egypt it takes an average of 549
days to register a new bakery. Levying taxes on unregistered
businesses is almost impossible, hence the
importance of making registration simpler. Simplifying
tax practices will compound the benefit.
Reduce taxes. Many emerging economies have generous
governments. But they fund their generosity by
imposing high taxes on companies in the formal sector.
This increases the unfair advantage enjoyed by
informal players, and puts them off crossing into the
formal sector. Lowering tax rates would tackle both
problems. Indeed, combining lower corporate tax rates
with stronger enforcement may well increase the overall
tax take.
Facilitate “creative destruction” in services.
Services are dynamic by nature. To maximize overall
service employment, companies must be free to start
up, grow, and create more jobs or—if they can’t compete—
to shrink, lay off workers, and close. To lubricate
this process of creative destruction, governments
need to make detailed policy changes.
Make it simpler to create and grow new firms, and
close failing ones. That means cutting the red tape
surrounding both business start-ups and bankruptcies. In
addition, governments should make it easier for small
businesses to prove ownership of their firms. Having
security of title means business owners can offer the
business itself as collateral for the loans it needs to
grow, and also sell it and move, when the right time
comes.
Enhance labor mobility. Labor laws intended to promote
job security and large severance costs deter firms
from taking on more people when business is brisk.
Firms even try to get round such legislation by employing
people as “temps” and then firing them just before
the law recognizes them as permanent employees. On
the other hand, restricting temporary or seasonal, or
part-time employment also makes it hard for businesses
to adjust staffing to fluctuations in demand.
Governments should examine their labor laws for such
unfortunate unintended effects, and revise them so that
employers can create jobs and workers can take them
more easily.
Local services have been left out of developing
economy growth strategies for half a century.
Import substitution, export manufacturing and,
more recently, services for export have captured policymakers’
imaginations instead. But dynamic, competitive
local services can unlock a huge contribution to
overall GDP growth and employment. In fact, achieving
higher productivity in local services is the only way
for middle-income—and developed—economies to
ensure lifetime employment for all. ◆
In South Korea, during its push for
manufacturing-led growth, banks were
prohibited outright from lending to
consumer service sectors like leisure
and real estate.