eScholarship UC Item 02f8z392

8/2/2019 eScholarship UC Item 02f8z392

1/13


2/13

Research ReportFisher Center for Real Estate and Urban Economics University of Cali fornia, Berkeley Fall 2003

The New Wave of Outsourcing

Ashok Deo Bardhan and Cynthia A. KrollThere is growing apprehensionamong business leaders, economists,and ordinary Americans that we arewitnessing what may well be thelargest out-migration of nonmanufac-turing jobs in the history of the US

economy. This concern has been fu-eled by newspaper reports and eco-nomic news highlighting the layoffsof thousands of people in high-tech,software and service sector compa-nies in the US, and the practicallysimultaneous, seemingly coordinatedestablishment of offices and devel-opment centers, most often in India,resulting in hiring of thousands ofnew employees in that country. Forexample, tabulation by the authors of

reports in Indian newspapers andbusiness journals for the month ofJuly 2003 alone gave an estimate of25,000 to 30,000 new outsourcingrelated jobs announced by US firms.In the same month, there were 2,087mass layoff actions carried out by USemployers resulting in a loss of226,435 jobs.1 The jobs being createdin India and elsewhere are in a widerange of services sectors such as geo-graphic information systems services

for insurance companies, stock mar-ket research for financial firms,medical transcription services, legalonline database research, and dataanalysis for consulting firms, in addi-tion to customer service call centers,

1 Bureau of Labor Statistics.

payroll and other back-office relatedactivities.

In this short overview we address thefollowing questions: Have jobs beentransplanted from the US? How sig-

nificant is this phenomenon and howsustainable is it? What is the potentialimpact on future job creation andwage inequality in the US? How is itlikely to impact the real estate sector?

The First Wave: Outsourcing ofManufacturingBetween 1987 and 1997, the share ofimports in inputs used in US manufac-turing increased from 10.5% to 16.2%and in high-tech manufacturing, such

as computers and electronics, from 26

to 38% (See Figure 1). These datacontinue a long history of foreign out-sourcing in US manufacturing and theassociated loss of blue-collar jobs inmany industrial sectors. Indeed, oneof the attributes of the modern stage

of globalization for advanced indus-trialized countries is the offshore pro-duction of intermediate inputs, usuallyin low-cost developing countries. Themotivation, on the part of US firms,has been driven by the low costs ofmanufacturing abroad, primarily inthe East Asian countries, such as Tai-wan, China, South Korea, Malaysiaand others, as well as the availabilityof skilled labor, the promotion of abusiness-friendly environment and the

existence of production and supply

Figure 1

Imported Inputs as a Share of Total Inputs

Total Manufacturing and High-Tech Sectors

0%

5%

10%15%

20%

25%

30%

35%

40%

1987 1992 1997

ImportedInputsas%o

fTotal

Inputs

Total Manufacturing High-Tech Manufacturing

Source: Bardhan,Jaffee and Kroll, Globalization and a High-Tech Economy,

forthcoming.


3/13

2

networks in those countries. At thesame time, the higher value-added,better paying jobs in management,finance, marketing, research and de-velopment have been retained in thehome country.

Considerable research has been car-ried out on the phenomenon of out-sourcing in manufacturing and manyof the economic insights and conclu-sions are applicable to Business Proc-ess/Services Outsourcing (BPO/BSO)as well. As pointed out by Bardhan,Jaffee and Kroll in their forthcomingbook, Globalization and a High-Tech

Economy, the outsourcing of parts ofthe supply chain of manufacturing has

resulted in a shift of demand, andhence jobs, from blue-collar to white-collar and from manufacturing to ser-vices, increased wage inequality be-tween blue-collar and white-collar

jobs, and increased profitability of USfirms. They also note that recession-ary downturns seem to prod firms intomaking major restructuring moves,and that a recession might be themother of innovation and dynamism.2

The New Wave: Outsourcing ofWhite Collar JobsThe software sector was the first ser-vice sector to transfer significant ac-tivity to foreign locations, leading tothe creation of a critical mass of ex-pertise and resources in concentratedlocales, such as the city of Bangalorein India. The rapid dissemination ofthe Internet, the transnational net-works set up by immigrants in the US,and liberalization of emerging market

2 Most economists believe, however, that

outsourcing should not lead to job loss inthe long run but to a reshuffling of jobsand a new composition of occupations in

the economy. This recovery of jobs lostto outsourcing still requires major changesin the industrial and employment structure

of the economy.

economies created the conditions for amajor burst of outsourcing in the1990s, in hitherto primarily domesticsegments of non-manufacturing sec-tors, such as telecommunications, re-tail trade, and finance (including

banking and insurance). While thepush factors for business processoutsourcing (BPO) or business ser-vices outsourcing (BSO) are similar tothose for manufacturing and arelargely cost-driven, the pull factorsand attributes of countries andeconomies providing outsourced ser-vices are somewhat different. In addi-tion to cost advantages similar tothose offered by the manufacturingcenters of East Asia, the ongoing out-

sourcing of business services jobs toIndia, Malaysia, Philippines andSouth Africa among others is also dueto the widespread acceptance of Eng-lish as a medium of education, busi-ness and communication in thesecountries; a common accounting andlegal system (at least in some of thecountries), the latter based on thecommon law structure of UK and US;general institutional compatibility andadaptability; the time-differential de-

termined by geographical locationleading to a 24/7 capability and over-night turnaround time; simpler logis-tics than in manufacturing, and asteady and copious supply of techni-cally savvy graduates.

Indias information technology en-abled services (ITES) sector, the pri-mary destination of business servicesoutsourcing from Western countries,now directly employs over 200,000

people with around $2.3 billion inexports, of which over 70% are to theUS. While the sector is still small it isgrowing at a rate of 60% per annum.The software services sector overallhas exports of approximately $9.5billion, of which over $7 billion are to

the US3. Indias National Associationof Software and Service Companies(NASSCOM), the primary trade or-ganization of all IT related firms,forecasts that exports would hit the$50 billion mark in the next five

years. By that time, the business proc-ess/ business services outsourcingsegment would employ over 2 millionpeople, and the total exports of the ITindustry would support over 8 million

jobs.

The growth of the IT sector in generaland the BPO segment in particular isnot confined to India. Firms involvedwith software services outsourcingand BPO are rapidly gaining ground

in the Philippines and Malaysia (callcenters and other back-office BPO),China (embedded software, financialfirm back-office BPO, some applica-tion development), Russia and Israel(high-end customized software andexpert systems), and Ireland (pack-aged software and product develop-ment). While it is difficult to estimatethe exact number of jobs created inthese countries in these sectors, letalone those transplanted and created

by US firms, tentative evidence col-lected by the authors suggests thatbusiness process outsourcing andsoftware outsourcing have togethergenerated, at the very least, over amillion jobs in the 1990s and hun-dreds of thousands more since the turnof the century.

BPO/BSO Impact on the USEconomyThe second half of the 1990s was a

time of high employment and robustgrowth for the software-related sec-tors, as well as the services sector atlarge. The job creation from outsourc-ing in countries around the world dur-

3 National Association of Software andService Companies, India, at

www.nasscom.org .


4/13

3

ing this period can be seen as spin-offs from the US because of tight la-bor markets, rather than job transfersout of the US in search of lower laborcosts. However, the recent downturnand the continuing jobless recovery

have legitimately given rise to thequestion whether services outsourcinginvolves the transfer of US jobs andoccupations to other countries. Table1 shows employment data for those

sectors of the economy that felt a dis-proportionate impact of outsourcing.These include the computers and elec-tronic products manufacturing sector(including its sub-sector, semiconduc-tors and electronic components); pro-

fessional and business services sectorssuch as business support services,which include call centers, and com-puter systems design services; andinformation industries such as tele-

communications, software publishing,and Internet services providers. Be-tween first quarter 2001 and secondquarter 2003, i.e. in the course of justover 2 years, the employment in thesesectors has plummeted by 15.5% in

the US as a whole, and 21% in thestate of California, corresponding to a

job loss of over 1 million and 200,000respectively in these sectors alone.

Table 1

Employment Change in Industries At Risk to Outsourcing*US Employment (Thousands) California Employment (Thou-

sands)

Industry Name Q1-2001 Q2-2003 % Change

2001-2003

Q1-2001 Q2-2003 % Change

2001-2003

Nonmanufacturing SectorsSoftware Publishers (except Internet) 276.1 247.9 -10.2% 55.8 47.1 -15.6%Internet Publishing and Broadcasting 50.6 33.7 -33.4%Telecommunications 1323.4 1138.9 -13.9% 150.5 123.5 -18.0%ISPs, Search Portals, and Data Proc-essing

516.0 433.2 -16.0% 60.2 48.0 -20.2

Data Processing and Rel. Services 320.9 292.2 -8.9% 24.4 18.9 -22.8%Accounting, Bookkeeping & Payroll 976.3 875.7 -10.3% 108.8 103.1 -5.2%

Payroll Services 158.9 124.6 -21.6%Computer Systems Design and Rel. 1341.2 1148.1 -14.4% 218.2 163.2 -25.2%Business Support Services 784.4 746.2 -4.9% 56.2 57.2 1.7%

Telephone Call Centers 406.2 363.2 -10.6%Telephone Answering Services 54.8 50.9 -7.1%Telemarketing Bureaus 351.4 312.3 -11.1%

Manufacturing Sectors

Computer and Electronic Products 1862.1 1415.9 -24.0% 443.1 336.8 -24.0%Semiconductors and ElectronicComponents

308.7 237.9 -22.9% 162.1 115.2 -29.0%

Subtotal: At-Risk Industries 6853.9 5791.8 -15.5% 980.8 774.6 -21.0%

All Nonfarm 131,073.0 130,515.3 -0.4% 14,608.2 14,491.8 -0.8%

Manufacturing 16,932.3 14,757.7 -12.8% 1,849.0 1,587.2 -14.2%Nonmanufacturing 114,141.3 115,757.7 1.4% 12,759.2 12,904.6 1.1%

* The authors have chosen those industries which, in our judgment, have been most often noted as outsourcing toIndia and East Asia. These industries have a substantial share of the occupations discussed in the next section.Source: Authors from US Bureau of Labor Statistics data.


5/13

4

Figure 2

Attributes of Jobs Outsourced

No Face-to-Face Customer Servicing Requirement

High Information Content Work Process is Telecommutable and Internet Enabled

High Wage Differential with Similar Occupation inDestination Country

Low Setup Barriers

Low Social Networking Requirement

Table 2Average Salaries of Programmers

Country Salary Range

Poland and Hungary $4,800 to $8,000India $5,880 to $11,000Philippines $6,564Malaysia $7,200Russian Federation $5,000 to $7,500China $8,952Canada $28,174Ireland $23,000 to $34,000Israel $15,000 to $38,000USA $60,000 to $80,000

Source: CIO magazine, November 2002, SmartAccess Survey, Merrill Lynch.

Indisputably, most of the job loss isdue to the technology downturn, thedot-com bubble, and the cyclicaldownturn in the US economy. How-ever, outsourcing that began as a re-sponse to very tight labor markets inthe US in 1999-2000 has continued,becoming a factor in the jobless orjob-loss recovery of 2003. As in thelast downturn in the early nineties,recession-based cost-cutting by firmsmay end up as the permanent loss of

jobs that remain abroad even duringthe subsequent recovery. The laid-offUS workers must then be absorbedeither in new sub-sectors, broughtabout by innovation, or in otherlesser-paying, non-tradable services

jobs.

Vulnerability to outsourcing extendswell beyond the sectors shown in Ta-ble 1. The employment services sec-tor, for example, lost over 300,000

jobs between June 2000 and January2001 and over 150,000 between Janu-ary 2001 and June 2003 (again a mixof recession-based losses and out-sourcing). Links to outsourcing in thissector come through temporary em-ployee agencies, which providedshort-term employees to many of theindustries listed in Table 1. Outsourc-

ing also has the potential to affect di-verse segments of retail and wholesaletrade, utilities and healthcare, to theextent that record-keeping, account-ing, sales, and information aspects ofthese sectors can be performed sepa-rately from other functions.

Outlook for ServicesOutsourcingThe occupational mix of a sector maydetermine its vulnerability. In

BPO/BSO circles it is said half-seriously that any job that involvesmostly sitting at a desk, talking onthe phone and working on a com-puter is a job under potentialthreat. Figure 2 summarizes the es-sential attributes and features of jobsand occupations that might find them-selves in jeopardy.

While institutional and cultural com-patibility and proliferation of the Eng-

lish language are key components ofcomparative advantage for countriesthat are destinations for BPO invest-ment and activity, it is the cost differ-ential, along with the availability ofwell-educated graduates, that providesthe critical competitive edge. As Ta-ble 2 shows, the salaries of computer

programmers in the emerging marketcountries of Asia and Eastern Europeare a factor of ten less than corre-sponding salaries in the US. The cost-differential in BSO is more difficult topin down, since the range of occupa-tions is so wide. Table 3 shows hourlywages for some sample occupationsfrom the July 2002 National Compen-sation Survey of the Bureau of LaborStatistics matched with comparableoccupations in India. The wage differ-

ential varies widely by occupation,with differences particularly high forlower wage, nonprofessional occupa-tions and less extreme, although stillquite significant, at the upper end ofthe wage spectrum.

A lower wage scale is even more at-tractive if it comes with a well edu-cated labor force. The three majoremerging market economiesChina,India, and Russiahave a sizeable

higher education sector. While Rus-sian expertise in many basic sciencesand engineering subjects has been

justly famous for decades, both theannual output and quality of science


6/13

5

and engineering graduates from Indiaand China have been increasing rap-idly and are now comparable to theadvanced countries4 (see Figure 3).These countries face some constraintsin exploiting this ongoing opportunity.Indias inability to provide educationat the basic school level could stiflefurther growth in highly trained

graduates. Russia faces growth con-straints from a combination of institu-

4 The figure for the US includes graduateswho are foreign citizens. However, the

proportion of foreign citizens is consider-able only at the PhD and MS level, not somuch at the basic undergraduate level of

higher education.

tional underdevelopment, erratic re-forms and the gradual deteriorationof the higher education system. Theoverpowering Chinese success inmanufacturing may well be repli-cated later in the services sectors, butas yet business services outsourcingfaces heavy language, institutionaland cultural barriers. Rising wages

and costs in these countries may spursecondary outsourcing to still less

developed countries, but from thepoint of view of the US labor marketsthat is no consolation.

Despite these barriers, the phenome-non of services outsourcing is sustain-able for the foreseeable future, unlessthere is a major disruption of the in-ternational economy or a severe back-

lash in the developed countries lead-ing to establishment of regulatory

Table 3Hourly Wages for Selected

Occupations

US and India, 2002/2003Occupation Hourly

Wage,US

Hourly

Wage,India

Telephone Op-erator

$12.57Under$1.00

Health RecordTechnologists/Medical Tran-scriptionists

$13.17$1.50-$2.00

Payroll Clerk $15.17$1.50-$2.00

Legal Assistant/Paralegal $17.86 $6.00-$8.00

Accountant $23.35$6.00-$15.00

Financial Re-searcher/Analyst

$33.00-$35.00

$6.00-$15.00

Source: US wages are from US Bu-reau of Labor Statistics, NationalCompensation Survey, July 2002; In-dia wages are from interviews, busi-

ness literature search and review ofemployment Want Ads by the authors.

Figure 3

Yearly Graduates with Natural Science

and Engineering Degrees 1998

0

50,000

100,000

150,000

200,000

250,000

Russia Ch

ina India

Japan

Germ

any

South

Kore

a US

DegreesGranted

Source: National Science Foundation (Science and Engineering Indicators, 1998).Note: Figures are by country where degree granted and may include foreign nationals.

Figure 4

US and California High-Tech Manufacturing Sectors

Annual Value-Addition Per Employee

40

60

80

100

120

140

160

180

1987 1992 1997

Thousandsof

dollarsperyear

US California

Source: Bardhan, Jaffee and Kroll, Globalization and a High-Tech Economy,forthcoming.


7/13

6

hurdles. The benefits to US firms arethe increased value addition and prof-itability resulting from savings due tolow-cost outsourcing. Figure 4 showsthe constant increase in value-additionper employee in high-tech manufac-

turing from 1987 to 1997, a period ofintense outsourcing activity in manu-facturing overall. The impact of thepresent cycle of BPO/BSO is perhapsreflected as well in the latest produc-tivity figures released by the US Bu-reau of Labor Statistics: Nonfarmbusiness output per hour worked in-creased by 5.4% in 2002, and by asizeable 6.8% in the second quarter of2003.

Outlook for US Jobs andOccupations If both the supply and the demandside suggest a sustainable outlook forbusiness services outsourcing, it isimperative to get at least a heuristic

sense of the potential size of the longterm impact on jobs and occupations.The authors have tried to arrive at anestimate of the outer limit of jobs po-tentially at risk to outsourcing byadopting the following methodology:

a) We focus not on economic andindustrial sectors, as in Table 1,but rather on the occupationalmake-up of the US economy,given by the detailed Occupa-tional Employment Statistics,2001, published by the US Bureauof Labor Statistics.

b) We are guided by the occupa-tional outsourceability attributeslisted in Figure 2.

c) We only take into account thoseoccupations where at least someoutsourcing has already takenplace or is being planned, accord-ing to business literature.

There are 22 broad occupational clas-sifications listed by the Bureau of La-bor Statistics.5 Within these 22 broadcategories there are 770 detailed oc-cupations. Table 4 shows the aggre-gate and detailed occupations which

we judge to be consistent with thecriteria a, b and c listed above. Ofcourse not all jobs are under threat inany of these categories. Table 4 liststhe outer limit of potential direct jobloss in these occupations, without tak-ing into account many of the dynamicadjustments that may take place orchanges that may occur in qualifica-tions, skill requirements and task de-scriptions.

Data on these occupations are avail-able for 2001 and some earlier years.The data indicate that these jobs spana wide range of compensation levels,from salaries one-third below the av-erage to almost twice the average sal-ary. In some outsourceable occupa-tions, job growth was strong at leastthrough 2000, but the occupationsmost vulnerable to outsourcing beganlosing jobs. For example, data entrypositions dropped by 115,000, or

22%, between 1999 and 2001, eventhough employment in computer oc-cupations as a whole was increasing.As occurred earlier in manufacturing,it was the lower paying, more routine

jobs that were being outsourced mostrapidly. This is consistent with theparticularly wide wage differentialsfound in the lower paying occupa-tions.

5 Many categories of these broad occupa-

tional classifications, such as personalcare and service occupations, foodpreparation and serving related occupa-

tions, construction, repair and mainte-nance related occupations, communityand social service occupations and others

are obviously non-outsourceable.

Table 4US Employment in Occupations at Risk to Outsourcing

Employment

Average Annual

Salary

Sectors 2001 2001All Occupations (Total US Employment) 127,980,410 $ 34,020Occupations at Risk of Outsourcing

Office Support* 8,637,900 $ 29,791

Computer Operators 177,990 $ 30,780

Data Entry Keyers 405,000 $ 22,740

Business and Financial Support** 2,153,480 $ 52,559

Computer and Math Professionals 2,825,870 $ 60,350Paralegals and Legal Assistants 183,550 $ 39,220

Diagnostic Support Services 168,240 $ 38,860

Medical Transcriptionists 94,090 $ 27,020

Total in Outsourcing Risk Occupations 14,063,130 $ 39,631

Percent of All Occupations 11.0%

Source: Authors using data from Bureau of Labor Statistics web site. *Officesupport aggregates data from 22 detailed Office and Administrative Supportcategories. ** Business and financial support aggregates data from 10 detailedBusiness and Financial Occupations. Further details on sectors available fromthe authors.


8/13

7

There have been many different esti-mates of potential job losses in the USfrom future business services out-sourcing. A report by Forrester Re-search forecasts that by the year 2015,approximately 3.3 million jobs will

have been irretrievably lost, almostone fourth of our estimate of total em-ployment in outsourcing occupationsat risk in 2001. This translates to alittle over 250,000 per year, a numberwhich seems conservative, based onthe rate of outsourcing over the lastfew years, the experience of outsourc-ing in manufacturing, the increasingability of an increasing number ofcountries to compete for these jobs,the higher tradability of services due

to better communications, increasinguse of English and US standards inbusiness and commerce, and the obvi-ous benefits to US firms and employ-ers, the primary decision-makers inthis process. This outsourcing of jobscould result either in net job loss insome occupations and sectors or in aslower pace of job expansion thanwould otherwise occur.

Outsourcing Has Regional

ImplicationsAs with manufacturing outsourcing,the process of services outsourcing islikely to vary geographically, amongdifferent regions of the US and withinmetropolitan areas. Figure 5 showsoccupations at risk for some of thelargest metropolitan areas in the US,while Figure 6 shows wage levels byoccupation, relative to the US, for thesame metropolitan areas. Most of thenation's large metropolitan areas have

a higher proportion of jobs in occupa-tions at risk than is found in the US asa whole, suggesting that many ofthese urban centers may share dispro-portionately in the wave of outsourc-ing. However, the occupational com-position of the at-risk jobs varieswidely among these MSAs, as dowage levels, and the type of job re

shuffling is likely to reflect these dif-ferences. Detroit has lower than aver-age shares of services jobs at risk tooutsourcing and may share less in theimpacts of this round of outsourcing(but has no doubt suffered frommanufacturing outsourcing in earlieryears). Atlanta has a high share of

office support occupations, at averagewage levels. Possibly an earlier re-cipient of jobs spun off from morecostly metropolitan areas, places likeAtlanta may be at risk of losing moreof their lower-wage outsourceable

jobs, although it could also continueto be the recipient of jobs outsourced

Figure 6

Salaries in Occupations at Risk of Outsourcing Relative to

Average US Salaries in At-Risk Occupations

2001, Selected MSAs

0.6

0.8

1.0

1.2

1.4

1.6

Atlan

taBo

ston

Chica

goDe

troit

Houst

on

NewYo

rk

LosA

ngele

s

SanF

rancis

co

SanJ

ose

1=SalaryatUSAver

agefor

Occupation

All Occupations Office Support

Business & Finance Support Computer and Math Professions

Source: Authors from Bureau of Labor Statistics data.

Figure 5

Occupations at Risk to Outsourcing as a Share of

Employment in All Occupations

Selected US MSAs

0%

5%

10%

15%

20%

US

Atlan

taBo

ston

Chica

goDe

troit

Houst

on

NewY

ork

LosA

ngele

s

SanF

rancis

coSan

Jose

PercentofEmploymentinall

Occupations

Office Support Business & Finance Support

Computer and Math Professions Legal and Medical



9/13

8

domestically from higher-wage areas.Within California, there is a widevariation among places. Los Angeles,with less than average shares of mostservices sectors at risk to outsourcingand close to average salaries within

these sectors, may have less to losefrom the next wave of outsourcingthan high priced markets elsewhere inthe state.

High-tech markets such as San Jose,San Francisco and Boston are particu-larly at risk of services outsourcingover the next decade. San Jose, theheart of Silicon Valley, has belowaverage shares of outsourceable officesupport and business and financial

support occupations, but almost fourtimes the average share of computerand math jobs (relative to its totalshare of US employment). At salarylevels well above the US average, theregion has already lost many of thelower-wage occupations to other partsof the country or abroad. Its vulner-ability now lies in the very high shareof high-wage outsourceable profes-sional occupations, many of which aresimilar to the types of positions grow-

ing in the lower cost foreign locationsdescribed earlier. Businesses thatforged a relationship with an overseassupplier at the height of the dot-comboom may continue to take advantageof the cost savings, despite the dot-com collapse and easing of demandfor these occupations in US locations.

Outsourcing has intraregional implica-tions as well, especially in the moremoderately priced urban areas. Some

of the largest overseas migrations ofservices jobs have been in occupa-tional categories that were once thecore of suburban job development,such as data processing and call cen-ters. Suburban locations that built up

an employment base of back officejobs could see these tenants shrink, orexpansion opportunities evaporate, asthese occupations shift overseas.

Present and Future Impact on

Office MarketsThe office building sector faces con-siderable uncertainty going forward.CB Richard Ellis reports that close to17% of for-lease US office space isvacant. Rosen Consulting Group

(RCG) figures show at least 700 mil-lion square feet are vacant in the of-fice-leasing market of major US met-ropolitan areas. Building activity inthe late 1990s, although more con-

strained than in the late 1980s, was

still the highest in a decade, as shownin Figure 7.

Because office construction tends toinvolve years of preparatory planning,much of the new space came on line

just as the dot-com bubble collapsedand employment in office-related sec-tors began to shrink. Employment inkey office sectors, on a national level,has dropped by 6.5% in the US and byalmost 10% in California since itspeak in 2000, in both cases returningto between 1998 and 1999 levels, asillustrated in Figure 8. The most vul-nerable sectors have been computer-related industries, telecommunica-

tions, and employment servicesthe

temporary employment services thathelped fuel the technology expansion.Many of these are the same sectorsnow undergoing extensive outsourc-ing.

Figure 7

Annual Value of Office Construction in Place

Constant Dollars, 1972-2002

0

10000

20000

30000

40000

50000

60000

70000

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

MillionsofDollars,1996base

Source: US Bureau of the Census.


10/13

9

Office vacancy rates respondedquickly to the combination of declin-ing employment and new space com-ing on line. Nationwide, rates dou-bled, from below 8% in December2000 to over 16% in June 2003, asshown in Figure 9. In California, va-cancies rose to an estimated 15.3%,ranging from below 10% in Sacra-

mento to above 20% in Silicon Valleymarkets.

Two factors are at work when va-

cancy rates risechanges in theamount of space occupied andchanges in the total amount of spaceavailable. Figure 10 shows occupiedand vacant space nationwide since1991, as distributed in downtown andsuburban markets (the four segmentsof each bar add up to total square

footage). Despite job losses due to arange of factors, the decrease insquare footage under lease (i.e. occu-pied) in suburbs and downtown areas

combined has been modestabout4% since 2000. The rest of the rise invacancy comes from a 6.6% increasein supply, which may come from new

construction or from existing build-ings entering the for-lease market (forexample, owner occupied buildingsmade available for lease). This sec-ond element of supply increase mayaccount for the difference between thepercent change in office employmentand the percent drop in space underlease. The new for-lease space added

to the total stock from owner-occupied buildings becoming for-lease buildings is actually a furthersign of declining demand.

Figures 9 and 10 also highlight a shift

that is occurring between suburbanand downtown areas. During the1990s, suburban markets led in netabsorption, and suburban vacancyrates dropped below downtown ratesfor most of the decade. By 1997, theamount of space vacant had shrunk tounder 350 million square feet in the73 markets tracked by RCG, with thevacant space almost evenly split be-tween suburban and downtown loca-tions. During the economic expansion

in the late 1990s, close to 85% of newconstruction occurred in suburbanareas, and downtown vacancies

dropped well below suburban rates.Suburban areas were hit much harderin the slowdown of 2001 and 2002.Several factors are likely at play. Onthe supply side, new suburban con-struction can proceed more readilythan infill development in many mar-kets. On the demand side, the typesof office occupations that have been

Figure 8

Employment Trends in Selected Information, Professional

and Business Services Sectors, US

0

2000

4000

6000

8000

10000

12000

14000

16000

1990

1992

1994

1996

1998

2000

2002

ThousandsofEmployees

Other Information Services

Telecommunications

Software+ISP/Web+Data

Proc

Employment Services

Consulting/Scientific R&D

Computer Systems Design

Arch/Design/Oth PST

Legal/Accounting/Advertising


Figure 9

US and California Office Vacancy Rates, 1996-2003

0%

2%

4%

6%

8%

10%12%

14%

16%

18%

20%

1996 1997 1998 1999 2000 2001 2002 Q2

03

PercentofSpa

ceVacant

US Metro

US SuburbanUS Downtown

CA Metro

Source: US data is from CB Richard Ellis. California data is from the authors,

as compiled from data from CB Richard Ellis, Cushman and Wakefield, Grubb

and Ellis, and Keegan and Coppin. Except for 2003, all years are for 4 th quarter.


11/13

10

outsourcing most rapidly have beenthose that are historically located insuburban areas.

Figures 9 and 10 may actually under-state the current vacancy situation inoffice space. Figure 10 includes bothunleased space and space available for

sublease in the vacant category, whilesome of the brokerage reports used forFigure 9 are less consistent and reportonly unleased space as vacant. Nei-ther chart takes into account buildingsthat have been taken off the market inthe most impacted areas because ofthe lack of leasing opportunities, orvacancies in owner occupied spacethat has not yet been offered for lease.

Growth in demand for office space in

the US will be tempered by a numberof factors of which services outsourc-ing is only one. Other factors includeunderutilized space currently underlease, the flexibility of square footageusage, and lessons in caution learnedfrom the recent boom. These factorsalso are likely to interact with oneanother. In markets already glutted

with space, some space is being heldoff the market, either in whole build-ings mothballed for the short term,or as empty space being held in an-ticipation of future growth in demand.These spaces could accommodate asignificant increase in demand with-out an apparent effect on vacancy

rates. As demand grows, firms thathave become more dependent on thebottom line may choose a more cau-tious route to space utilization than inthe last expansion, making more effi-cient use of existing space before tak-ing on obligations for additionalsquare footage.

Outsourcing will further dampen thegrowth in demand for space, andcould even lead to declining demand

in some markets. The Forrester Re-search estimate of 3.3 million jobs isequivalent to between about 500 and800 million square feet of office space(depending on the ratio of square feet

per employee)6at the higher end

6 The ULI Office Development Handbook

reports industry standards at 200 to 250

surpassing the amount of space cur-rently vacant in for-lease buildingsnationwide. Not all of these jobs arein sectors heavily present in for-leaseoffice space. Nevertheless, manytypes of office markets could feel the

effects of outsourcing. Those most atrisk may be back office suburbanmarkets in slow growth or decliningmetropolitan areas, but the high-techmarkets that are just beginning to re-cover from the dot-com bust may alsofeel the effects of the occupationalrestructuring that comes with servicesoutsourcing.

Concluding RemarksThe US economy underwent a major

wave of outsourcing in manufacturingindustries, a process that gatheredmomentum in the 1980s and 1990sand continues today. The experienceof that phenomenon provides a usefulbenchmark for evaluating the currentwave of outsourcing in the servicessectors. Business process and businessservices outsourcing will have a sig-nificant impact on the economic land-scape in the US. Several major differ-ences distinguish services outsourcing

from the previous wave of outsourc-ing of manufacturing jobs. Servicesoutsourcing is structurally simplerthan manufacturing outsourcing interms of resources, space and equip-ment requirements and thus may pro-ceed much more quickly. Servicesoutsourcing affects overwhelminglywhite-collar middle class jobs andoccupations, unlike manufacturingoutsourcing, which impacted primar-ily blue-collar workers. In addition,

this time around it is a different set ofcountries that are in contention forthese jobs. Figure 11 summarizesthese differences and their implica-tions for the economy.

square feet per employee, but also notesthat in some markets the ratio may be as

low as 150 square feet per employee.

Figure 10

Occupied and Vacant Office Space

US Major Office Markets, 1991-2003

0

500

1,000

1,5002,000

2,500

3,000

3,5004,000

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003f

MillionssofSquareFeet

Suburban Occupied Downtown Occupied

Suburban Vacant Downtown Vacant

Source: Authors from data supplied by various brokers to Rosen Consulting Group.


12/13

11

While our report has focused primar-ily on the US economy as a whole, theeconomy of California is equally vul-nerable. As seen in Table 1, the statessectors at-risk to outsourcing havefared more poorly in the last two anda half years, than the US average . Interms of future impact, bear in mindthat while the state does not have too

many of the call center and data entrylevel type jobs anymore, it has aheavy presence of the computer re-lated occupations, as well as office,legal and healthcare support jobs.Moreover, the cost differential withthe rest of the world is higher, thussuggesting a higher incentive for jobmigration abroad. Finally, large num-bers of temporary foreign employees,such as computer engineers from In-dia in large California based firms,

sensing the way the wind is blowing,have requested within-firm transfersto subsidiaries in their home coun-tries.

While evidence from the recession ofthe early 1990s suggests that a majorbenefit of globalization has been thegrowth in high-tech services employ-

ment that accompanied the outsourc-ing of manufacturing production, it isnot clear how the economy will adjustto the present burst of services out-sourcing. At least four different out-comes are possible.

One possible scenario is that servicesjob outsourcing proves more costly to

the economy than the earlier round ofmanufacturing outsourcing. As cen-ters of skilled high-tech professionalsbuild up in other parts of the world,the US and California may no longerdominate the next wave of innova-tions, and we would observe slowergrowth of high-wage jobs within theUS and California. In this extremesituation, economic adjustment, in theabsence of continuing innovationoriginating in the US, first might take

the form of prolonged unemployment.Then, workers losing their jobs to out-sourcing would finally be absorbed inlesser-paying services jobs. Alterna-tively, there could be a downwardadjustment of salaries and wages,making the outsourced occupationsinternationally competitive again. Un-der this worst-case scenario, the im-

pact on the demand for office spacewould initially be reflected in lowerrents and prices, and higher vacancyrates. In the long run, with increasingemployment in other jobs and occupa-tions, rents and prices would settle on

a lower growth path trajectory withvacancy rates returning to their long-run equilibrium.

As an alternative to this troubling sce-nario, a backlash against globalizationcould occur, both worldwide andwithin the US, slowing down theprocess of business services outsourc-ing. Opponents of globalization arealready discussing protectionist meas-ures and regulatory roadblocks in the

form of restricting the kind of jobsthat can be outsourced. If successful,this kind of protectionism, althoughinefficient from the point of view ofthe economy, may result in the reten-tion of some of the outsourceable

jobs. In the short run, this wouldmoderate the negative impact on thereal estate sector.

A third possibility is that the industryshrinkage shown in Table 1 may come

in part from domestic outsourcing,indicating a redistribution of jobswithin the US rather than a net loss.This could involve vertical disintegra-

tionthe shifting of jobs from largeemployers to smaller firms in supportsectors--as well as the ongoing proc-ess of domestic outsourcing fromhigh-cost regions such as California torelatively low-cost regions elsewherein the United States.7 This processwould mitigate the differences in

prices and rents among different re-gions within the nation and wouldleave the nationwide vacancy and ab-sorption rates relatively unaffected.

7 To the extent that the outsourced work isdone by start-up firms, employment num-bers may currently be undercounting the

current employment situation.

Figure 11

Impact of Outsourcing

Impacts blue-collar jobs

Affected individual industrial

sectors and some specialized

occupations within them

Job losses offset and even

reversed by increases in

services employment

Led to increased inequalitybetween blue-collar and

white-collar occupations

Impacts white-collar jobs

Affects individual occupationsin many industrial sectorsacross the economy

May lead to differentcomposition of occupations inthe economy; unclear how thelabor market adjustment willwork.

Will lead to increasedinequality within white collaroccupations

Manufacturing Services


13/13

12

Rents in some of the higher pricedmarkets could continue to remain de-pressed, even with expanding em-ployment nationwide.

Finally, the most positive scenario is

that the US and California economiescontinue to fashion their outsourcingactivities in light of the new produc-tion paradigm, keeping the cream ofthe new development at home, while

the more routine activities are out-sourced. Under this scenario, innova-tion would lead to a continuing streamof new service and manufacturingactivities, and hence new jobs andoccupations, while competition and

the need for lower-cost supply wouldforce more mature services operationsoverseas. Depending on their educa-tion and skills, individual workersmight still find it difficult to find re-

placement employment at similarwages, but overall, the jobs lost tooutsourcing would be replaced byhigher-wage jobs in the new sub-sectors brought about by innovation.Increasing wages, incomes and com-

pany profits would then impact thereal estate sector positively through arecovery and eventual increases inprices, rents and occupancy rates.

Ashok Deo Bardhan is Senior Research Associate and Cynthia A. Kroll is Senior Regional Economist at the Fisher Cen-

ter for Real Estate and Urban Economics. Further information on outsourcing trends in high-tech manufacturing and

services sectors and more generally on globalization and the h igh-tech economy is available in their forthcoming book,

Globalization and a High-Tech Economy, coauthored with Professor Dwight M. Jaffee, Willis Booth Professor of

Banking, Finance and Real Estate at the Haas School of Business and Co-Chair of the Fisher Center for Real Estate

and Urban Economics.

eScholarship UC Item 02f8z392

Documents