The World Bank
DECEMBER
ECoNOMIC POLICY
[[@)<l [)g2@SS11180C8$2 2 2 34
Export processing zones are potentially useful tools for export promotion. To
foster development, however, zones must be set up properly, managed well,
and integrated with other reforms.
Over the past three decades export pro-  spillover, demonstration effects, and back-
cessing zones have become popular instru-  ward linkages.
ments of trade policy, offering firms located  Zones generally share several common
in them free trade conditions and a lib-  features:
eral regulatory environment. In 1970 only  o They allow duty-free imports of raw and
ahandful ofcountriespermitted such zones,    intermediate inputs and capital goods  As export
but by 1996 there were more than 500 zones    for export production.
in 73 countries (OECD 1996). This note   o Government red tape is streamlined,  processing zones
outlines the general features and objectives    allowing "one-stop shopping" for per-
of these zones, highlights country experi-    mits, investment applications, and the   have become more
ences with them, and offers policy recom-    like. In addition, labor laws are often
mendations for establishing them.        more flexible than for most firms in the   widespread, the
domestic market.
ITMV  s a gon                          o Firms in zones are given generous, long-  policies governing
An export processing zone is one of many    term tax concessions.
export promotion tools, including bonded   o Communications services and infra-  them  have become
warehouses and temporary admission       structure are more advanced than in
schemes. Today's export processing zones  other parts of the country. Utility and   more important
have evolved from their original defini-    rental subsidies are common.
tion as "an industrial estate, usually a fenced-  o Zone firms can be domestic, foreign,
in area of 10-300 hectares, that specializes  orjoint ventures. Foreign direct invest-
in manufacturing for export." (World Bank  ment plays a prominent role.
1992, p. 7). Many firms, called export pro-    Two main features differentiate export
cessing firms, now benefit from the incen-  processing zones, however. First, zones can
tives offered in the zones without being   be publicly or privately owned or managed.
physically fenced in. In addition, coun-  Overthepastl0-15yearsthenumberofpri-
tries have liberalized domestic sales-Mex-  vately owned or managed zones has grown
ico, for example, allows 20-40 percent of  substantially because they are believed to
its zones' output to be sold domestically.   achieve superior results. Second, zones can
Export processing zones have three main  be "high-end" or "low-end," depending on
goals. They are to provide a country with  the quality of the management, facilities, and
foreign exchange earnings by promoting  services they provide firms.
nontraditional exports. They are to create
jobs and generate income. And they are  /Zna erpaRr_eS
to attract foreign direct investment, engen-  Not all export processing zones have served
dering technology transfer, knowledge  as engines of industrialization and growth,
FROM THE DEVELOPMENT ECONOMICS VICE PRESIDENCY AND POVERTY REDUCTION AND ECONOMIC MANAGEMENT NETWORK



as some proponents anticipated (Warr 1989,  also occurs at the supervisory and managerial
1993; World Bank 1992). Zones have been   levels, with local employees establishing for-
engines-among others-in the economy   eign contacts and learning new organiza-
when they have been given their proper place  tional and managerial methods, negotiation
as a policy tool, andwhen their ultimate costs  and marketing skills, general business know-
and achievements are taken into account.  how, and a spirit of entrepreneurship.
Zones have been very successful in Mauritius,  Catalyst and demonstration effects are also
for example, but have failed in Senegal.  common in the host economy (Rhee, Kat-
Export processing zones are sensitive to  terbach, and White 1990; Rhee and Belot
the national economic environmentand per-  1990), though backward linkages between
Properly managed,    form better when the host country pursues   zones and domestic firms have not always
sound macroeconomic and realistic exchange   occurred. In general, some linkages have
zones can generate    rate policies. Properly managed, however,  occurredincountriesthatdidnotalreadyhave
zones can generate income and createjobs-   a solid industrial base. But these linkages have
income and                 especially nontraditional employment and   been spotty and inconsistent, with some zone
new employment opportunities for women   firms complaining of the poor quality or
create jobs                (table 1). Their long-term contribution is  incompatibility of local inputs. (Exceptions
twofold. A competitive, efficient zone pro-  include the Dominican Republic and Mau-
vides an industrial infrastructure that many   ritius.) Linkages have also occurred in
countries lack. In addition, zones build   economies-the Republic of Korea, Taiwan
human capital directly and through their cat-  (China)-tiat had a solid industial base prior
alystanddemonstrationeffectsonhostcoun-  to the establishment of the zones. In these
try entrepreneurs.                         cases, however, export processing zones were
Export processing zones build human cap-  only one of the tools used to foster growth.
ital in two ways. Previously unskilled work-  Zones have increased gross foreign
ers benefit fromjob training and learning by   exchange earnings. In Mauritius zone export
doing (Rhee, Katterbach, and White 1990).  earnings jumped from 3 percent of gross
These benefits are limited, however, because   export earnings in 1971 to 67 percent in 1995.
most production is low-skill and low-tech.  Net foreign exchange earnings have not
Still, workers earn income and learn indus-  always lived up to expectations, however. In
trial work discipline and routine. Training   1996 gross exports of threeJamaican zones
Table a  Employment creatlron and ezos in selezc&    eorp  pocessing zones
Region                        Year of         Year          Number          Number             Zone exports!
and economy                 first zone      of data         of workers      offirms         gross exports (percent)
Africa
Cameroon                      1990            1995             2,567             16                       32.5
Mauritius                     1971            1995            80,466            481                       67.0
Senegal                       1974            1990              600              10                        1.8
Asia
Bangladesh                    1983         1995-96            37,533         2 zones                      11.1
Philippines                   1972            1994            81,559    from 4 zones                      11.0
1991           43,858             -                   7.7 (1990)
Sri Lanka                     1979            1998            85,323         6 zones                       -
1990           60,000                                      23.0
Taiwan (China)                1966            1995            55,191    235 (3 zones)                      5.6
Latin America and the Caribbean
Costa Rica                    1972            1996            24,000            191                       21.0
El Salvador                   1976            1991             6,500             14                       14.5
Honduras                      1976            1991            19,000    49 (9 zones)                      12.0
Jamaica                       1976            1996            16,804     56 (3 zones)     16.2 (2.0 of net exports)
Source: World Bank data.



totaled $235 million, but net exports were  port credit insurance), an export processing
just $29 million. Such earnings may not cover  zone is a second-best policy choice.
a country's investment in a zone. Similarly,    Zones can play a long-term dynamic role
the opportunity cost of such investments may  in their host country's development if they
not warrant the concessions granted (on   are set up properly, managed well, and inte-
income taxes and tariffs) to accommodate  grated with a national reform and liberal-
a zone. In fact, some zones may have a neg-  ization program. At the very least they should
ative net present value for the country.  not become impediments to reform.
Wages in most export processing zones   For three reasons zones should not be
are equal to or higher than average wages  established in liberal, low-protection
outside the zones, though there is consid-  economies. First, lower than expected or  Zones should
erable variance around this mean. For  unsatisfactory foreign direct investment
instance, in Thailand in 1990 and Sri Lanka  may be due to inadequate laws or regu-  not be
in 1992 wages in export processing zones  lations, or to other distorted economic
were higher than outside the zones. But in  incentives (such as weak private property  established in
Mauritius in 1986 and Taiwan (China) in  laws). Second, zones are distortionary
1988 they were lower. And lax labor, safety,  trade instruments and introduce an ele-  liberal, low-
and health laws in many zones have raised   ment of discretion in the policy environ-
concerns about workers' welfare.        ment. Finally, even if export promotion is  protection
The environmental impact of zones and   appropriate (and compatible with World
lax regulation and monitoring have also raised   Trade Organization guidelines), an export  economies
concern. Environmental pollution has been   processing zone may not be the best way
confirmed in some zones (for example, in  to achieve that goal. If post-reform eco-
the Dominican Republic and Mexico). But  nomies are intent on establishing new
systematic analysis is lacking that would lead  zones, differential fiscal incentives should
to well-targeted, sensible regulation and mon-  be minimized to reduce their distortionary
itoring. Some observers argue that govern-  impact on the host economy.
ment attitudes toward worker and health
conditions in and the environmental impact  Policies for success
of zones may not be much different from   An export processing zone is more likely to
prevalent national practices. Still, these are  succeed when monetary and fiscal policies are
important concerns.                    sound and stable, private property and invest-
Some analysts consider a successful zone   ment laws are clear, firms are free to repatri-
a model for host country policymakers to  ate earnings at market rates, and there are
mimic in liberalizing the rest of the econ-  no restrictions on foreign exchange.
omy. Others argue that a successful zone  Zone firms should be moderately taxed.
may become a stumbling block for liberal-  There is no need for overly generous tax
ization by providing jobs and foreign  incentives (permanent tax holidays, waiv-
exchange earnings, and thus easing the pres-  ing of all taxes). Indirect taxation and licens-
sure on policymakers to undertake econo-  ing should be rationalized and minimized,
mywide reforms (as in India and Tunisia).  and zones' imports and exports should be
Other countries (Uganda) have considered  free of trade taxation and tariffs.
or established zones after macroeconomic  Utilities (water, electricity, sewage, and the
and trade reforms to bolster foreign direct  like) should not be subsidized; doing so dis-
investment.                             courages economicallyrational use of resources
and factors ofproduction, underminingzones'
General recommendations                 benefits for host countries. Providing infra-
In distorted economies an export processing  structure outside the zone-telephony, roads,
zone is one of a number of tools used to off-  ports-can have positive spillovers for the local
set anti-export bias. Butjust like other export  and national economy by facilitating trans-
promotion tools (bonded warehouses, ex-  portation and communications. But onsite



infrastructure-pavement, building shells-  lic, considerable autonomy should be
should be privately financed.            granted (as in Taiwan, China).
Business-friendly labor laws lower labor  Zone firms exporting from one mem-
costs. But strengthening regulations and   ber of a trade arrangement should be aware
monitoring can mitigate potential labor law   of potentially complex rules of origin and
infractions and improve working conditions,  restrictions.
reducing turnover and absenteeism and
increasing worker productivity.          Further reading
An essential first step toward minimizing   Madani, Dorsati. 1998. "A Review of the Role
environmental impacts is to develop a qual-    and Impact of Export Processing Zones."
Governments               itative and quantitative understanding of    World Bank, Development Research
industrial refuse and its effects on air, soil,  Group, Trade, Washington, D.C.
should be fully           water, and human health. Follow-up regu-  OECD (Organisation for Economic Co-oper-
lations, incentives, and monitoring should  ationandDevelopment). 1996. Trade,Employ-
appraised of the          be tailored accordingly.                    ment and Labour Standards: A Study of Core
Zones in countries that are members of   Woriers'Rights and International Trade. Paris.
costs and benefits        preferential trade arrangements (regional or  Rhee,YungWhee, andThereseBelot. 1990.
bilateral) may be more attractive to firms  Export Catalysts in Low-Income Countries:
of the incentives         targetingthese markets, because such amem-  A Review of Eleven Success Stories. World
bership enlarges potential market size and  Bank Discussion Paper 72. Washington,
they offer to export    easesentrybarriers (asinMauritius).Exports    D.C.
from these zones may, however, face complex   Rhee,YungWhee, Katharina Katterbach, and
processing zones          rules of origin regulations and restrictions.  Janette White. 1990. "Free Trade Zones
in Export Strategies." Industry Series Paper
Administration, regulation, and            36. World Bank, Washington, D.C.
incentives                               Warr, Peter G. 1989. "Export Processing
Governments should be fully appraised of the  Zones: The Economics of Enclave Man-
costs and benefits-to the budget and to the  ufacturing." The World Bank Research
country-of the incentives theyoffer to export  Observer4 (1): 65-88.
processing zones. Incentives should be com-    . 1993. "The Potential for Export Pro-
patible with World Trade Organization rules  cessing Zones: Lessons from East Asia."
and timelines on export promotion instru-  Pacific Economic Bulletin 8 (1): 19-26.
ments; otherwise hostcountries mayface retal-  World Bank. 1992. "Export Processing
iatory actions by importing countries. In  Zones." Policy and Research Series Paper
addition, zones and export processing firms  20. Industry and Energy Department,
should be allowed to locate in various loca-  Washington, D.C.
tions (as in Mauritius).
Government should provide efficient,  This notewas written byDorsatiMadani (Consult-
streamlined, and prompt services for set-  ant, International Trade, Development Research
ting up and running export processing   Group).
zones (approval of investment applications,    If you are interested in similar topics, consider
customs and other supervisory institu-  joiningtheExport Competitiveness Thematic Group.
tions) .Privately owned and managed zones   Contact GanyPursell, x38002, orcickon Themnatic
should be encouraged. If zones are pub-  Groups on PREMnet.
"^  i t This note series is intended to summarize good practice and key policy findings
on PREM-related topics. PREMnotes are distributed widely to Bank staff and are
h      lD   also available on the PREM website (http://prem). If you are interested in writing
a PREMnote, email your idea to Asieh Kehyari. For additional copies of this PREM-
NO       w  note please contact the PREM Advisory Service at 87736.
Prepared for World Bank staff