POLICY RESEARCH WORKING PAPER   , 3.=
Enterprise Restructuring                          Tec  !puI.
in  Eastern  Europeha.prudms
!.            Sbs i',
How Much? How Fast? Where?                            c~n~Id5sn
Preliminary Evidence from  Tra-de                        ts.onre
Data
W    "' t-' ""un   t'46
~-~he~maime gam iyJ.. 
Bernard Hoekman                                       -*.-..
ar<gdthi e2Ot9.~ ,. .>
Gerhard RPobetrutrn<
Thereecm Wooreld Bank
Africa Regionsnd 
TecHnical Much?p HowartmentWh
aWa~~~~~~~~~~~~~~~~~~~~~~~~~~ ..--<;^.
Finance and Private Sector Development Division
March 1995                                              -..-,.



PLicy RESEARCH WORKING PAPER 1433
Summary findings
What kind of privatization program is best suited to     either are not available or are unreliable.
stimulate enterprisc restructuring in former centrally     The only objective measure available - comparable
planned economies? One view, expressed by mobt           across countries - is export performance. Trade data
Western business and political leaders, is that          reveal to what extent firms have been able to reorient
privatization is best pursued case by case, with emphasis    themselves to create and exploit competitive advantages.
on sales to new owners, including foreign investors.     We now have four years of data, cnough to get an idca
Another view, espoused by a minority of "radical"      of what is going on, and to comparc one Central or
economists and economists-turned-politician, was that    Eastern European country's performance against
restructuring is best pursued through economic           another's.
incentives; combined with "mass" privatization of state    The data suggest that the Czech Republic, the country
enterprises so that they become widely held (public) joint    chat has pursued mass privatization most actively and
stock corporations.                                      credibly, has also done best in restructuring its industries
The ultimate test, of course, is future productivity   and reorienting them toward world markets. Those that
growth and rising welfare standards. We cannot yet       pursued a gradualist approach - Hungary being the
measurc these. The disaggregaped data on production      main example - have changed their export structure
and employment by industry required for such a measure   less, but export growth has also been above average.
This paper- a product of the Finance and Private Sector Development Division, Europe and Central Asia, and Middle East and
North Africa Regions, Technical Departmcnt - is part of a larger cffort in the region to analyze progrcss-and policy options in
transition economies. Copies of this paper are availablc free from the World Bank, 1818 H Strect NW, Washington, DC 20433.
Please contact Ms. Faten Hatab, room H8-087, extension 35835 (2S pages). March 1995.
The Picy Resarch Workig Paper ers dissemnafes the fing of work w progress to aecowage the exchang  of ies aboua
deilpment isus An objectie of the seris is to get the fndings ot qrlry, eu if dtie presentatios are kss dtn fidly polshed The
paper cary the ames of the authors a should be used anadcco&dngy. The find cinterpwtatons, and conclusions are the
,ahort own and sh    not eb ahRbuted to he World Bank its Exeaiue Board of Directrs, or ay of its member courtne&
Produced by the Policy Research Dissemination Center



Enterprise Restructuring in Eastern Europe
How Much? How Fast? Where?
Preliminary Evidence from Trade Data'
Bernard Hoekman and Gerhard Pohl
The World Bank
We are very grateful to Ying Lin for his excellent research assistance in compiling the
database used in this paper; to Alexander Italianer, Jerzy Rozanski, Zhen Kun Wang and Alan
Winters for their help in obtaining and understanding the raw data; to Faten Hatab for report
processing and preparation; and to Roland Eggerer and participants in a World Bank seminar for
helpful comments.



1.    Introduction                                  The data suggest that the country that
pursued mass privatization the most act;vely and
What kind of privatization program is  credibly--the Czech Republic-has done the best
best suited to stimulate enterprise restructuring  in terms of restructuring its industries and
of formerly centrally planned economies? This    reorienting them towards world markets. Those
question has been hotly debated since 1989.  that pursued a gradualist approach-Hungary
Two opposing views have been most prominent.   being the main example-have unuergone less
The first view, expressed by most Western     change in the structure of exports. However,
busiiness and political leaders is that privatization    export growth has also been above-average. The
is best pursued case-by-case, with emphasis on  two countries where the output collapse has been
sales to new owners including foreign investors.    most severe-Bulgaria and Romania-have the
The sezond view, espoused by a minority of    weakest overall trade performance. Bu'lgaria
'radical' economists and economists-turned-   differs from Romania, however, in that the
politicians was that restructuring is best pursued    composition of its ecports to the EU changed
through economic incentives, combined with    very rapidly, suggesting that substantial re-
"mass" privatization of State enterprises to  structurinig is occurring.
become widely held ('public') joint stock
corporations. Hungary, for example, has              Shifts in the structure, composition and
followed the gradual, case-by-case approach; the    geographic pattern of trade flows alone are, of
Czech Republic and Russia have relied on mass    course, i'ot necessarily indicative of restructuring
privatization. In Poland, mass privatization has    at the leve-l of the firm. Firms could have
been held up for nearly four years.           reoriented exports from Eastern to Western
countries without altering any of the character-
The ultimate test, of course, is future  istics of the goo-is involved, and therefore
productivity growth and rising welfare standards.   without restructuring. However, CMEA and
We cannot yet measure these. The              Western markets were to a large cxtent seg-
disaggregated data on production and          mented, with different quality standards and
employment by industry required to do this are  designs. Except for raw materials, the shift
either not available or unreliable.' The only  towards Westem markets is not possible without
objective measure available, comparable across  considerable 'restructuring' by adopting new
countries, is export performance. Detailed    product design, technologies, quality control,
export statistics provide not only an indirect  marketing strategies, and so forth. Restructuring
measure of structural change, but export per-  may or may not involve visible actions such as
formance has a very direct impact on produc-  plant closures and layoffs.
tivity, since higher exports will permit higher
imports of scarce Westem capital goods and           Aggregate unemployment may not
services (computers, machinery, process con-  embody much information on the magnitude of
trols, telecommunication equipment, software,  restructuring that is occurring in an economy. It
management consulting, etc.). In the near term,   has sometimes been argued that the Czech
productivity growth will crucially depend on  Republic's unemployment rate of only 3.5
such imported items.                          percent indicates that much of the required
restructuring is yet to come, especially as rates
Trade data reveal to what extent firms  vary between 10 and 16 percent in the other
have been able to reorient themselves to create  CEECs. The export performance of the Czech
and exploit competitive advantages. We now    Republic is at odds with this view.2
have four years of data, enough to get an idea of
what is going on, and to compare the                 The impact of the European Union's
performance of the Central and East European  (EU) trade policy vis-a-vis the CEECs has been
Economics (CEECs) to one another.            just as hotly debated as privatization.3 Regmlts to
date suggest that exports have not been



2
Figure 1: Per capita exports, 1989 and 1993 (US$)
ToaEU              ToTbhe World
1000
900
800                  "I 
700
600                       H      Lry
-500      CR      R
400
300  l   Lria ri                         d
200                    ~ 
100 -
0
89 93   89 93   93 93   89 93   89 93   89 93    899    89es93
constramined by restricted market access for        The various measures that are discussed
'sensitive' products. Rapid growth in exports  in what follows are indirect indicators of the
occurred in both sensitive and non-sensitive  restructuring underway. What matters is not the
items. Indeed, export growth in certain sensitive   absolute value of any indicator, but how thz
products-agriculture foremost-has been far   Eastem European countries compare to each
below potential (market access granted),     other. Such comparisons provide some
suggesting that the CEECs may not have as    information on the impact of the different
strong a comparative advantage in agriculture as   strategies with regard to 'restructuring' policies:
is often believed.                           privatization, trade liberalization, etc. Macro-
economic factors will of course be important as
well, but mainly affect aggregate performance,
2.    Measures of Change: Aggregate    not changes in the structure and composition of
Trade Indicators                       exports. It is the latter that is the focus of
attention here.
What follows presents and discusses a
number of sunmnary measures that have been   Table 1: Share of EU in total exports.
compiled on the basis of import data reported by
the European Communities. Data on the imports
from the CEECs are used rather than reported   CEECs                  29.3       52.3
exports by each of the CEECs.4 We also mostly                           .        32A
abstract from exports to the rest of the world.
Although this reflects data availability, an
analysis that is limited to the EU is of interest   It is useful to start with a brief
given the fact that the CEECs are clearly    discussion of total exports of the CEECs and the
.specializing' in the EU market. The EU now  FSU in the last 5 years. Figure 1 reports per
accounts for over 50 percent of the CEECs'   capita exports of the CEECs. to the world and to
exports (Table 1).                           the EU. Two points stand out. First, the
decline in the value of total external trade after



3
1989, due to the demise of the CMEA was offset  their share of the EU market, and Bulgaria
by 1993 for the majority of the CEECs.5  Per  almost doing so.
capita exports from thc CEECs to the EU nearly
doubled. The growth rate of per capita exports    Table 3: Average annual growth in exports
to the EU is the highest for the Czech and            to EU
Slovak Republics,' at 26 percent per year on
average, followed by Bulgaria and Poland at 20                       1989-93 (%)
percent, and Hungary with 14 percent. The      Bulgaria                   16
absolute value of per capita exports to the world  CSFR                  24
is highest for the Czech Republic at $1,000    Hungary                    11
followed by Hungary at around $850 in 1993.    Poland                     18
Second, a marked shift in exports towards the  Romania                   -10
EU occurs. Of the comparator countries, only
China achieved similar performance.' The       FSU                         4
Czech and Slovak Republics also demonstrate
one of the largest shifts in the geographic
pattern of exports, by 30 percentage points.'       The Czech and Slovak Republics are the
'star' performer on the basis of growth in
The CEECs' trade structure has changed   exports to the EU (Table 3). In 1989 reported
substantially in the last four years. Total  imnports by the EU from the Czech and Slovak
exports to the EU rose by 14 per cent per year  Republics, Hungary and Romania were roughly
oii average between 1989 and 1993 (from ECU  equal at ECU 2.5 billion. By 1993, exports
12 billion in 1989 to 20 billion in 1993). In this   from the Czech and Slovak Republics had
period, total EU imports rose by only 2.1 per  attained ECU 6 billion, Hungarian exports were
cent per year, while world trade increased by  slightly below ECU 4 billion, and those of
3.7 per cent per year. The CEECs therefore   Romania had fallen to ECU 1.7 billion
substanitially increased their market share in the    (Figure 2). Exports of the Czech and Slovak
EU (Table 2).                                Republics grew by 24 percent per year on
average during the 1989-93 period,
Table 2: Market share in the EU 1%)          outperforming China. Poland and Bulgaria
follow second and third in terms of export
1989       1993     grcswth.
CEECs                  2.7        4.1
FSU                    3.0        3.2      Table 4: Actual and 'long run' predicted
Bulgaria               0.12       0.19              exports to EU (ECU billion)
CSFR                   0.56       1.23
Hungay                 0.58       0.31
Poland                 0.85       1.54                          1993       Predicted
Romania                0.57       0.34      Bulgaria             0.9            3.9
CSFR                6.0           12.5
Hungary             3.9            4.5
Poland              7.5            8.5
For the six countries that have signed  Ronania             1.7            3.0
Association Agreements-Bulgaria, the Czech     FSU                15.5           31.2
Republic, Hungary, Poland, Romania and the
Slovak Republic-total import penetration rose
from 2.7 per ce:it in 1989 to 4.1 per cent in
1993. Individual coumtry market share growth
was often substantially higher with both Poland     How significant are these changes in
and the Czech and Slovak Republics doubling  trade patterns? Spatial trade pattems are usually



4
Figure i  Exports toAU, I 986-93               ratios will be influenced by valuation effects that
affect GDP calculations (e.g., exchange rate
levels), they are nonetheless suggestive. The
0                                         Czech and Slovak Republics and Bulgaria appear
, - -   to be the most 'open', significantly more so than
the other CEECs. This in tum has implications
for the potential of these countries to absorb
0. 0                  imports of investment goods which, in tum, will
6                , Aolmd       ,         strengthen the restructuring process. A
-          - &SR           comparison with the long-run openness ratios
,'  /             predicted by a gravity model (Hamilton and
5       -"            /                   Winters, 1992) suggests that the CEECs appear
,            o                    to have come close to attaining the future
.4  ,                                     (Figure 3). The main exception is Hungary, in
it   /   ,, -  li' uniya;'  part due to its (less undervalued?) exchange rate.
3                                         Figure 3: Actual and 'predicted' openness
, - ,                                          ratios (exports/GDP)
2    N
1  1   .                                     40
_~~~~~~~~~~~~~~~3
1909    1990    1991    1992    1993
30
25
well explained by 'gravity' models where trade
between partner countries is determined by the
GDPs of the two partners, their populations, and
the geographic distance between them. They can    Is
be used as one benchmark against which the
trade performance of the CEECs can be             10
compared. If the actual value of exports in 1993
are compared with a gravity model-based            s
'prediction' of longer-run 'potential' exports
(Baldwin, 1994), the CEECs have already            0
bitguia CP&SR Humggly utiaad Riumk FW
moved substantially towards realizing their            *   2.450   a    2.270  1.090  1.ooe
potential (Table 4).                                             F njiaMPII
The sanme conclusions emerge if actual  *, WwrlBI At
EU market shares and export-to-GDP ratios are
compared with gravity model predictions              Trade performance of the last 4 years
(Appendix Table 1). The 'openmess' of CEEC     suggests that the long run will be achieved much
econoniies as indicated by their export-to-GDP  more rapidly than most observers expected in
ratios varies from 16 percent for Poland to 35  1989, and the composition of exports
percent for Bulgaria (Figure 3). Although these    substantially different than what was predicted.



5
For example, Aghion et al. (1992) concluded  Flgure 4: Structural change In exports
that elimination of barriers to imports by the EU        relative to sample mean,
would increase exports of Czechoslovakia,               1989-93
Hungary and Poland by 40 percent. In the
event, exports increased by 95 percent during    F1pl-n C                Uin
1989-93. Agricultural exports were expected to   3owth .. °hph wr
expand by 75 percent; they actually fell slightly.
Mastropasqua and Rolli (1994), using the      10* 
SMART partial equilibrium simulation model,    srn 
predicted that EU liberalization would increase  .-.
'traditional' exports in goods with export values  .0.
above $3 million (at the 8-digit level) in 1988 by
$245 million, $175 million and $115 million, for
Poland, the Czech and Slovak Republics, and    s*%
Hungary, respectively. In fact, exports of such  e0.
goods increased by $850 n'.llion for Poland and  _0%
$215 million for the Czech and Slovak          ,0-
Republics, and declined by $220 million for    20%
Hungary.9 As discussed at greater length below,  10%
the importance of 'traditional' exports had fallen  oI
dramatically by 1993.                         .40%
4 0;
3.    Disaggregated Measures of                40X
Change in Export Composition                           n  Hunu Ptmd RAS u   U  aFMd
Restructuring occurs at the level of the
firm. The trade data that have been discussed so   expressed as a function of the mean change in
far are highly aggregated and only suggestive  trade structure for the sample of selected
regarding the magnitude of the changes that have   countries.'" As can be seen in Figure 4, the
been occurring at the finn/industry level.   structural changes in exports are much higher for
Although trade statistics are not reported at the    the CEECs than for other countries. This
level of the finn, detailed statistics are available,   suggests substantial restructuring is occurring.
for about 10,000 product categories imported by    The absolute value of the change between 1989
the EU. These data provide a valuable source of   and 1993 is particularly high for the Czech and
information on the changes in production for  Slovak Republics." The relatively high number
export in the CEECs and the FSU.             for Bulgaria reflects in part the small initial
value of trade with the EU.
A useful indicator of the change in the
structure of exports is the sum of the absolute     As mentioned earlier, the change
changes in the value of exports of individual  indicator gives equal weight to positive and
commodities between 1989 and 1993, divided by   negative changes. Both are equally relevant as
the 1989 value. This measure is reported in  far as restrucring is concerned. Figure 4 also
Appendix Table 2. It treats increases and    shows that the Czech and Slovak Republics have
decreases symmetrically, and is intended to give    very few declining export sectors: over 80
an impression of 'how much' change occurred in   percent of the dhange reflect commodities
exports. For ease of comparison, the figures are   where export growth occurs. Romania lies at



6
the other end of the scale: only 30 percent of  Figure 5: Relative Importance of 'expired'
the change is due to growing exports.                      and new products Iparcent)
If rcstructuring is occurring one would 
expect to observe that some products cease to be         Li"' t     plud US tiR01 Nin|
exported while other 'new' products come on       4
stream. Figure 5 shows that the CEECs greatly     36
expanded the number of 8-digit tariff line items  30
exported to the EU during 1989-93. The Czech      25
and Slovak Republics stand out in particular in   20
this regard, increasing the number of lines       15 i
exported from 41.6 percent of all the goods       *0
imported by the EU in 1989 to 61.5 percent in 
1993. All the CEECs have rapidly diversified       6 _X
their exports to the EU. Comparator countries     .5
have managed to diversify into new products      .10
only at a much slower pace. With the exception   -15
of China, the number of iterns exported to the   .20
EU by comparators such as Israel or Turkey is    .25
now significantly less than the Czech Republic,   dO
Hungary and Poland.                               .35
-40
BIua&  CRISR Huigmy  Ptimd  bmuxan  FU
Figure 6: Number of commodities exported               How large are the exports of the 'new'
to EU, 1989 and 1993                          owlreaeteeptsfth'n '
t  commodities relative to total exports? The value
j 1989 0 1993                of exports of new and 'expired' tariff lines
relative to total exports provides some
6500                                          infornation in this regard. 2 New items
61  5                           constitute a significant share of total exports in
Iow |        l l                              each year of the 1989-93 period (Appendix
550l                                          Table 3). Figure 6 provides a summary view of
|0W1                                          the relative importance of 'expired' and 'new'
|4^500 8     l l  l l                48.0     products in total exports. Two measures are
4000          43.   411                       reported: (i) the share of products that ceased to
* |  1   1 1   1 E   | |  |  be exported between 1989 and 1993, expressed
3500                                          as a percentage of 1989 total exports to the EU;
3000   3                       28.     |      and (ii) the share of 1993 total exports to the EU
2500 -                     9       5.that comprise products that were not exported at
2000 -19.                   20.               all in 1989.
1 Soo                                                The value of conmmodities that cease to
1000                                          be exported tends to be smaller than the value of
Soo -                                         new products that come on stream. New
1     o 1                        _,         products account for between 10 and 30 percent
Womb C SR  Hung"   P        Iamm   FSI   of total exports in 1993 for the CEECs. Expired
Nowl Minimum  .70o:1prinam tql, toamdmw ammbuum  products tend to represent the equivalent of 10-



7
20 percent of total exports in 1989. They     Tabl 5: Share of 'traditional' products 1' In
exceed by a substantial margin the changes in         total exports, 1989-93 (percent)
exports by comparator countries such as China
or Turkey. Of the CEECs, Bulgaria is the most               Share In  Share In   Annual
'dynamic'. Over the four year period, the                     19C9 1993    Average
equivalent of almost one-quarter of 1989 exports                              hi Value
by value 'disappears', while about one-third of  Bulgaria        44       19       -5.6
1993 exports are 'new' in the sense that products  CSFR          60       30       4.1
are involved that were not exported in 1989.
The numbers for the other CEECs are somewhat     hungary         59       35       -2.1
lower, but still high.                           Poland          74       50       7.2
Ronania        80       51      -19.4
Another way of getting a sense of the     FSU             96       80       -0.8
changes that have taken place is to look at the                                        l
performance of 'traditional' or 'established'  !'  Greater than ECU 3 min in 1989
exports since 1989. As noted earlier,
commodities that were important export items in   particular shows very large shifts towards
1989 have become much less significant (see  leather/footwear; fibers; raw materials/cement;
Table 5). By contrast, 8-digit product categories   and nonferrous metals, and away from steel and
that accounted for less than ECU 3 million in  annaments."4 Romania also reveals large
1989 grew extremely fast in all the CEECs.    changes: towards leather/footwear and garments,
Average annual growth rates for this category  and away from mineral resources andnonferrous
were 43 percent for the Czech and Slovak      metals. The Czech and Slovak Republic
Republics; 40 percent for Poland; and 55 percent   demonstrate a sharp shift away from agriculture,
in the FSU (Appendix Table 4). The share of   wood products and arms, and towards
this group of goods in total exports consequently   nonferrous metals, transport equipment and
increased sharply. For the Czech and Slovak   machinery. Poland becomes specialized in wood
Republics such 'non-traditional' exports      products and transport equipment, and less
accounted for 70 percent of total exports in  specialized in agriculture. Hungary shows the
1993, as compared to only 30 percent in 1989.    least change of the CEECs, in part due to its
For Bulgaria, the figures are 80 and 20 percent,   longer exposure to world market prices.
respectively. This illustrates the importance of
decentralized production and trade decision          The expansion of the number of new
resulting from privatization and price        products and the absolute number of tariff lines
liberalization.                               exported to the EU suggests that much of the
trade that is occurring between the EU and the
The change in specialization of the    CEECs is intra-industry trade, that is, the same
CEECs in their exports to the EU that occurred    types of goods are shipped both ways. Intra-
over the 1989-93 period is an additional source    industry trade has been expanding very rapidly
of information.'3 Specialization indices for  between the CEECs and te EU. Intra-industry
broad categories of exports to the EU in 1989  trade intensity indices for the Czech and Slovak
and 1993 are reported in Appendix Table 5.    Republics and Hungary now exceed the value of
The CEECs tend to be specialized in leather   intra-industry trade for countries such as South
products and footwear, garments, ferrous and  Korea or Israel, and are approaching levels
nonferrous metals, and furniture. Bulgaria,   registered by the U.S. or the EFTA countries in
Hungary and Poland are also specialized in   their trade with the EU.I5 Intra-industry trade is
agriculture, although their relative specialization    currently at about the level it was for trade
in this sector has been declining. Bulgaria in  between North and South Europe in the -arly



8
1980s (when it varied beiWeen 0.4 and 0.6, sec  Table 6: Share of 'subcontracting' In total
Nevci, 1990). Much of the East-Wcst intra-             exports of the EU  M%)
industry trade is with Germany and the EU's
Southern members (Neven, 1994).                                        1939    1993
Bulgaria           4.4    13.7
The rapid increase in intra-industry trade    CSPR               5.1    12.6
has implications for adjustment costs and market    lHungary           16.4    20.2
has mpliatios fo adjstmet cots ad makeLPoland              9.1    18.6
access opportunities in the futuire. Adjustment      Romania            12.9    30.1
costs abroad are likely to be lower in an intra-     CEEC               10.4    17.9
industry trade context becanse jobs lost due to      FSU                0.8      1.9
import penetration may be offset by job-
enhancing expansion in demand for imports from
the foreign partner for simi:ar goods. The    to obtain mnarket access, as trade policy in the
political opposition to liberalizing and expanding    EU tends to be less restrictive for processed
intra-industry trade is generally much more   goods reflecting a perception that it allows high
muted than in instances where trade flows are  cost industries located in the r U to subcontract
predominantly of the inter-industry type.'6   labor-intensive parts of their production process
to third countries, thereby allowing them to
Intra-industry trade can also be a      remain in business. Duties on goods that are re-
mechanism through which transfers of          imported after processing are usually based on
technology occur. The Europe Agreements with   the value added abroad, not on the gross value
the CEECs have increased the incentives for EU    of the goods. The Europe Agreements
su'npliers/retailers to engage in so-called outward   eliminated tariffs on such trade (Naujoks and
processing trade, where raw material or       Schmidt, 1994). However, it may also reflect
components are exported for assembly in lower    imperfections in the policy and/or regulatory
wage countries and then re-imported. Outward  environment in the processing country.
processing has been used intensively for sectors    Subcontracting can allow foreign partners to
such as gannents, electrical machinery and    trade with local finns without having to establish
furniture. Importers provide designs, insist on  production facilities. Of course, this is only the
quality control, take care rf mnarketing, etc.  case insofar as technologies are not such that
This is a good way for firms in partner countries   control through equity investment (FDI) is
to reduce the costs and risks associated with  required. The reach of such subcontracting is
development of export markets, while at the   therefore limited to relatively 'low-tech'
same time obtaining know-how from suppliers.    activities. Subcontracting can nonetheless do
Outward processing trade can be seen as a form  much to employ local factors of production and
of non-equity foreign direct investment.      transfer general foreign know-how and
management practices.
Export under outward processing
arrangements accounted for about 18 percent of
total CEEC exports to the EU in 1993, up from  4.    Expanding and Declining
10 percent in 1989 (Table 6). Most of the            Products
processing occurs in leather/footwear, clothing,
electrical machinery, precisic i instruments and     The combination of expansion in the
fumiture. All of these activities are labor    product line' exported to the EU and the decline
intensive.'                                   in the relative importance of 'traditional' exports
has led to the establishment of a more diversified
Outward processing trade may be driven    export base in the CEECs. This Section
by a variety of forces. It may be used as a way



9
provides a flavor of the dynamism demonstrated    cement, prefabricated buildings, metal furniture,
by entreprencurs in the CEECs and thc FSU by    ammonium nitrate, ceramic building bricks,
focusing on some of the fastest growing and  Iypewriters, DC motors, compact discs,
dcclining products in their cxport trade. The  upholstered seats and parts thercof, electronic
response to the changes in the incentivc structure   components, tantalium, footwear, electric
rcsulting from price liberalization, privatization  energy, sea-going vessels, live cows, doo,s and
and vr.atcr acccss to the EU markets has been  windows, reservoirs, builders joinery, raw hides,
nothing less than impressive.                sewing machines, and articles of plastic. The
ave- Be annual growth rate of exports of these
What are the growth eroes?                  products was over 150 percent, an impressive
performance by any standardl20
Clothing, footwear, and machinery dominate in
all the CEECs. Exports of these products grew  Hungary
very fast in the 1989-93 period. Box I provides
a listing of the top ten 'new' products-defined    Hungary's dynamic exports to the EU
as being zero or very close to it in 1989-   are similar to those of the Czech and Slovak
exported to the EU in 1993, sorted by absolute  Republics. Machinery and textiles and clothing
increase.                                    are the most important exports. Hungary differs
from the Czech and Slovak Republics in that a
Bulgaria                                     number of agricultural exports (sunflower seeds
and sausages), organic chemicals, and wood
For Bulgaria, clothing and footwear   products are products that have grown rapidly.
accounted for a quarter of all exports to the EU,   As can be seen from Box 1, agricultural
up from less than 10 percent in 1989. The    products figure prominently in Hungary's top
second major growth area for Bulgaria are silver   'new' exports.
and copper which increased from virtually zero
percent to 9 percent of total exports. Products    Hungary's overall trade performance has
where exports are both substantial and grew by  been weaker than that of the Czech and Slovak
over 100 percent per year include cathodes,  Republics. As noted earlier, aggregate exports
disodium carbonate, frozen fatty goose and   to the EU g:ew by 'only' 11 percent per year on
duck, conveyors and transmission belts, ceramic   average during 1989-93, as compared to 24
tiles, and bovine leather.                   percent for the Czech and Slovak Republics.
The number of dynamic 8-digit level product
Czech and Siovak Republics                   categories is consequently lower than for most
other CEECs.
Exports of machinery and vehicles
dominate export growth in the Czech and Slovak   Poland
Republics. The largest category in terms of
absolute growth was motor vehicles, followed by    Clothing exports grew the most,
automotive components, fixed electrical      accounting for 17 percent of total exports in
capacitors, motor vehicle parts' and footwear.  1993, up from 9 percent in 1989, followed by
Growth rates of many products are frequently  furniture and wood products exports 12 percent.
triple digit, and sometimes approach quadruple  The relative importance of transportation
digits.                                      equipment (cars, ships, aircraft) doubled to 10
percent of total exports.
Products for which exports grew by
more than 100 percent (from already substantial    At the 8-digit level, motor cars, pordand
base)19 per year during 1989-93 include: portland   cmxent and upholstered seats are the mostrapidly



10
Box I
-Tog 10 'New Exwor Products by Countrv (8-digdt level) and Value In 1993 (ECU million)
Bulgaria   Unrefined copper             22.2    Poland     Sea going vessels        124.3
Unwrought silver             18.2                Civil aircraft          53.8
Uppers of leather            16.9                Frozen fillets of fish  52.0
Women's footwear             9.8                Article of iron and stcel   31.2
Wire of refined copper       9.6                Discharge lamps          25.6
Special spirits              6.9                 Live domestic cows      23.2
Fatty livers of duck         5.8                Women tronsers           19.6
Transmission belts           4.6                Meat and edible offal    18.8
Turbo jets                   4.0                 Lignite (unagglomerated)  18.7
Gas oils                     3.8                Mens footwear with sole  17.8
CSFR       Fixed electrical capacitators    33.9    Romania   Mens footwear with sole  17.6
Unwrought tantalum           31.5               Ash and residues         11.7
Uppers of leather            30.1               Single phase AC motors   10.3
Parts/accessories of motors  26.7               Sea-going tankers        9.7
Motor cars                   24.8               Women trousers           8.5
Articles of iron and steel   24.4               Meat and offal of cows   7.3
Mixtures of ammonium nitrate  24.1               Fuel oil from bitumen   6.5
Machinery and appliances     22.8               Mens footwear with sole  6.4
Mens footwear with outer sole  19.5             Ignition wiring sets     4.7
Electrical energy            19.2                Flat rolled products    4.5
Hungary   Reciprocating piston engines   45.1
Mens footwear with outer sole  28.2
Fatty livers of domestic geese  27.5
Lambs, less than one year old  27.2
Meat and edible meat offal   26.2
Parts of electrical machinery    25.3
Reciprocating piston engines    25.2
Women footwear with sole     18.8
Articles of iron and steel   15.0
Parts of motor vehicles      14.9
Live domestic cows           14.3



11
expanding exports. Items with triple digit   Table 7: Russia coming on stream
growth rates (and more than ECU 10 million in         (ECU bn)
1993) include sea-going vessels, civil aircraft,
cathode ray tubes, gravel, cement clinker,                            1992        1993
certain chemicals, leather, ignition wiring sets,  Non-oil expons      2.6         5.4
copper wires, rails, ferro-chromium, cooking
appliances, removable insoles, reservoirs, zinc,
french windows, beech wood, paper, and       Declining products
builders joinery.
The summary measures of change in
Romania                                      exports discussed earlier embody 'sunset' as well
as 'sunrise' industries. Identification of the
Romania's exports fell substantially over   declining products in exports to the EU also
the 1989-93 period, largely reflecting the   provides information on the extent of
collapse of exports of mineral fuels, aluminum,    restructuring in the CEECs and the FSU. The
furniture and wood products, and iron and steel.   hardening of budget constraints and privatization
programs should result in large declines in
Significant growth occurred in exports of   exports of those goods that used to be either
clothing and footwear. Their share in total  heavily subsidized and/or were revealed to be
exports to the EU rose from 16 to 45 percent in    nonprofitable. There is quite some diversity
1993. There are few significant and rapidly  across the CEECs in this regard. The discussion
growing 8-digit items besides clothing and   that follows uses more aggregated data (2-digit)
footwear.                                    than the foregoing Section.
FSU                                                 The CEECs can be separated into two
groups: those with a large number of product
Most of the growth in exports from the  categories where exports to the EU decline; and
FSU are natural resources or natural resource-  those with only a small number of declining
based products. Copper and silver are two    products. Bulgaria, Hungary and Ramania are
major exports with growth rates exceeding 100  in the first group, the Czech and Slovak
percent per year on average. Other dynamic   Republics and Poland in the second. The FSU is
products are flat-rolled steel products, sunflower   somewhere in between. Agricultural products
seeds, bovine leather, ammonium nitrate, waste    and foodstuffs figure prominently in the
and scrap of nickel and stainless steel, and  declining export categories for the Czech and
collectors pieces (art works). There are few  Slovak Republics, Hungary and Poland.
clothing or machinery products in the most
dynamic 8-digit items. However, more recently    Bulgaria
manufactured exports have been expanding
similarly rapidly as in Central Europe. As          One third of product groups exported to
shown in Table 7, non-oil exports by Russia to    the EU had negative growth rates, of which two-
the EU more than doubled between 1992 and    thirds fell by 10 percent per year or more. The
1993, the year after price liberalization.   largest absolute decline occurred for iron and
Footwear and clothing exports have been      steel exports. In contrast to the other CEECs
growing very fast-often at triple digits-rising  (see below), agriculture plays a relatively minor
from ECU 9 million in 1989 to ECU 295        role in the declining product groups. Items
million in 1993, with most of the growth after  where large declines occurred include soaps,
1992.                                        wickerw Ac, wadding and felt, books, silk,
nickel products, e'ssential oils, and fish.



12
Czech and Slovak Republics                   Romania
Major declining exports (average annual      Declines in exports to the EU are
growth rates of minus ten percent or more) are    reported for more products than for any other
concentrated in a0riculture. Half of all products    CEEC. Mineral fuels dominate the decline in
where exports to the EU declined are         absolute terms, falling from ECU 844 million in
agricultural. Fish, preparations of vegetables,  1989 to ECU 37 million in 1993. There are
dairy products, and meat are among the major  only two agriculture items among the 12 worst
declining commodities. The largest absolute  performing products, dairy and meat. Both of
declines occurred for dairy and meat. Non-   these experience a dramatic fall in exports,
agricultural products where exports fell     diminishing over sevenfold between 1989 and
significantly include lead, ores, and wood pulp.    1993. Manufactured goods among the worst
performers include paper and paperboard; man-
Hungary                                      made fibers; motor vehicles; wood products;
iron and steel; furniture; and aluminum.
Products that decline by 10 percent or
more per year include vegetable saps, milled  FSU
agricultural items, nickel products, cement, fish,
fur, wool, ores and slag, fibers, fats, food        The largest absolute declines in exports
residues, dairy products, and fertilizers. The  to the EU occur for wood pulp, moto. vehicles,
largest absolute declines were registered for  wood products, and nickel products. Exports of
fertilizers and meat (each down by over 80   these four items fall by almost ECU 1 billion
percent of 1989 exports); and iron and steel  between 1989 and 1993 (equal to about half of
(down by 30 percent of 1989 exports). As for  1989 exports). Declines registered for many of
the Czech and Slovak Republics, agricultural  the products where exports are large tend to be
goods and processed foodstuffs figure        relatively small, however. Products where
prominently in the decline in exports. Between    exports were almost flat include spirits, rubber
1989 and 1993 exports of declining agricultural    articles, machinery, and organic chemicals.
items fell from ECU 600 million (one quarter of
total exports) to ECU 475 million (12 percent of
total exports to the EU).                    5.    Sustainability of Export Growth
Poland                                              One of the striking developments in the
structure of exports of the CEECs duriLng the last
About half of the 2-digit commodity   four years has been the convergence in
groups registering declines are agricultural or  specialization of exports to the EU that appears
foodstuffs. The latter account for most of the  to have taken place. Correlations between
absolute reduction in exports experienced by the    specialization indices for countries in Eastem
declining products: they were ECU 229 million  Europe are now quite high and have been rising
lower in 1993 than in 1989, accounting for 85  where they had been low in 1993 (see Table 8
percent of the declining group total. The only  and Appendix Table 6).  As of 1993,
non-food related item -<here exports fell by a  correlation coefficients between any of the
la&e amount is starlies and glues (down by 50  CEECs were at least 0.2, and for some country
percent of *91-4 exports).                   pairs were close to or above 0.5. Exports
patterns are thus becoming increasingly similar.
The same is true for the FSU, but from a lower
base and a more recent starting point.



13
Table 8: Relative specialization:
correlation coefficients, 1993       6.    Conclusions
BUL  CSR   HUN   POL  ROM                Trade data suggest that a significant
Bulgaria   1.00                             amount of restructuring has been achieved in the
CSFR     0.21  1.00                         CEECs in the last four years. Although total
Hungary   0.35  0.28   1.00                 exports of these countries as a group have barely
Poland    0.28  0.52   0.60  1.00           attained the levels of 1989, a massive
Romania  0.24  0.45   0.33  0.49   1.00     reorientation of trade towards the EU has
FSU      0.20               0.19            occurred. Given the need to produce to high
Western standards, while remaining price
competitive, this reorientation must reflect
restructuring at the firm level firm. This may or
An obvious question suggested by this  may not involve significant shedding of labor. It
finding-and the export performance of the     also does not necessarily imply large changes in
CEECs more generally-concerns the importance   the relative specialization of the economy. The
of comparative advantage as opposed to trade  CEECs demonstrate large differences in the
policy. What role have the Europe Agreements  extent to which changes in relative specialization
played?2' The fact that the CEECs now have    occur, and in the magnitude of the shifts away
preferential access to protected markets (e.g.,  from 'traditional' and towards 'new' products.
footwear, textiles and clothing) may be a
determinant of what we observe. At the same          The trade data that have been discussed
time, quotas and tariffs on sensitive items may  in this paper are perhaps most useful for cross-
induce a shift towards non-sensitive products that   country comparisons. Figuri 7 summarizes the
have been granted free access into the EU.    main message that is suggested by the available
data. It plots the change in per capita exports to
Although the CEECs are specializing in    the EU against the measure of change in the
many products where comparator countries have   'structure' of exports to the EU over the 1989-
been confronted with contingent protection, the  93 period. The Czech and Slovak Republics
growth in the relative importance of intra-   clearly outperform the other CEECs and the
industry trade, and the ongoing               FSU by a substantial margin.
diversification/differentiation of industrial
exports--especially from the Czech Republic and      While all the CEECs have managed to
Poland-suggests that the CEECs may have less    increase their exports to the EU very
to fear than their develoning country competitors   substantially in the last four years (Romania
in this respect.' The CEECs will gradually    being the exception), the Czech and Slovak
adopt EU laws and practices relating to       Republics taken together are the most successful
competition, thereby reducing the scope for EU  in this connection. In absolute ternis, exports
industries to invoke antidumping.23 Diversion of   have grown the most, and they also dominate
pressure for contingent protection may well turn    with respect to the number and importance of
out to be more important. Given the similarity    'new' products in total exports. At the same
between the CEECs and developing countries,   time, the Czech and Slovak Republics report the
the latter may come to face a more hostile trade    fewest negative-growth products.
policy environment in the future. Clothing
producers in particular will face fierce             The large increase in the number of
competition from the CEECs in the ne,.t 5 years    products exported to the EU and the growth in
as their preferential access to the EU is     intra-industry trade suggests that a lot of
maintained/strengthened.



14
Figure 7: Change in per capita exports to EC and Structural change in exports
2.00
1.80                                                * CR&SR
1.60                      * Bulgaria
CD                                 a Poland
c1 A0                  China
.  1.20              -              CEEC  * Hungary
ia 1.00
; 0.80          0 RoiWOsu
L- 0.60
0.40
0.20
0.001                                      I            I
-100          0           100         200          300
Change in per Capita Exports to EC. 1989 - 1993
restructuring is happenning through product        One would thus expect that the earliest
differentiation- Although aggregate measures of   reformers would have seen the largest change in
relative specialization have not changed very  trade structure during the four year period 1989-
much for the Czech Republic, there have been  93. This is indeed the case, if one compares the
significant changes within the broad product  CEEC average with the FSU average. But this
groups that were already exported to the EU. A    is not the case in Central and Eastern Europe.
similar conclusion holds for Hungary and    The Czech and Slovak Republics show much
Poland, but Hungary's aggregate performance  more change and growth than Hungary and
has been lagging.                           Poland. This could be either attributed to factor
endowments, macro-economic policy, or
It is, of course not possible to prove  different privatization strategies. All three
beyond all doubt cause and effect of industrial  probable played a role: Hungary had the most
restructuring, revealed by these remarkable  open economy before the reforms, and can thus
changes in export structure. But the data are,  be expected to show somewhat less change. It
nevertheiess, suggestive. Among the CEECs, all  has also not used exchange rate adjustment as
countries have liberalized prices of manufactured   extensively as the other countries to increase
goods early on in the refonn process (between  export competitiveness. But it possibly does not
1989 and 1991). Rapid changes in export     matter all that much as wages are only 10%
structure and performance generally started  (Poland), 11% (Czech and Slovak Republics) or
immediately following price reforrn (see    15 % (Hungary) of German wages, just across
Appendix, Table 2), for example, in Hungary  the border.
and Poland in 1990, in former Czechoslovakia in
1991. and in most of the FSU in 1993.



15
Since Polish wages are not all that   References
diffcrent from Czech and Slovak wage levels,
credibility of privatization and reform strategies    Aghion, Philippe, et al. 1992. 'Towards and
must have clearly played a major role. The  Integrated European Continental Common
same is true for Hungary. It's better starting  Market," in J. Flemming and J. Rollo (eds.),
point can explain the lesser change in export  Trade, Payments and Adjustment in Central and
structure. But wage differentials cannot account   Eastern Europe. London: Royal Institute for
for all the difference in export performance  International Affairs.
(Figure 7), and the fact that the Czech Republic
has quickly by-passed Hungary as the most    Baldwin, Richard. 1994. Towards an
successful exporter (Figure 1).             Integrated Europe. London: CEPR.
The conclusion can be sumomizd as     Borensztein, E., D. Demekas, and J. Ostry.
follows: mass-privatization has certainly    1993. 'Output Decline in the Aftermath of
stimulated at least as much (if not significantly  Reform: Bulgaria, Czechoslovadia, and
more) industrial restructuring than case-by-case    Romana," in M. Blejer et al. (eds.), Eastern
privatization with it's emphasis on "ideal"  Europe in Transition: From Recession to
owners who have the restrcturing know how.  Growth?. Discussion Paper 196. Washington
Technicians and managers, brought up under the   D.C.: The World Bank.
old system, have proved to be quite successful in
learning new management and marketing skills.    Greenaway, David and Robert Hine. 1991.
Some pressure on all enterprises led economy-  "Intra-Industry Specialization, Trade Expansion
wide to much larger changes than extensive  and Adjustment in the European Economic
restrucuring of only a few large firms. Foreign    Space," Journal of Common Market Studies, 29,
investment (initially concentrated in Hungary)  603-22.
has so far not had a large impact. Some
restrucuring has occurred everywhere even   Halpern, Usl6. 1994. "Comparative
before privatization had been implemented. The    Advantage and the Likely Trade Pattern of the
collapse of central auftority (Bulgaria, FSU) has   CEECs," CEPR Discussion Paper 1003.
been of econonic benefit in the short run.
However, credible and comprehensive         Hamilton, Carl and Alan Winters. 1992.
privatization and reform strategies are necessary    "Trade with Eastern Europe," Economic Policy,
to ensure a sustained restructuring effort. The  14:78-116.
reforms in the Czech Republic remain the
standard by which other countries will have to  Hoekman, Bernard and Petros C. Mavroidis.
be measured.                                 1994. "Linking Trade and Competition Policies
in Central and Eastem Europe," Policy Research
Paper 1346, The World Bank.
Kminskd, Bartomiiej. 1994. "The Significance
of the 'Europe Agreements' for Central
European Industial Exports," Policy Research
Worldng Paper 1314, The World Bank.
Mastropasqua, Cristina and Valeria Rolli. 1994.
"hdustial Countries' Protectionism with



16
Respect to Eastern Europe," T7he World
Economy, 17:151-69.
Messerlin, Patrick. 1993. "The EC and Central
Europe: The missed rendezvous of 1992?," 7he
Economics of Transition. 1, 89-1 10.
Naujoks, Petra and Klaus-Dieter Schmidt. 1994.
"Outward Processing in Central and East
European Transition Countries," Kiel Working
Paper 631 (May).
Neven, Damien. 1990. "Gains and Losses from
'1992'," Economic Policy, 10:13-62.
Neven, Danien. 1994. "Trade Liberalization
with Eastern Nations: How Sensitive?," CEPR
Discussion Paper 1000.
Pohl, Gerhard and Istvan Dobozi. 1995. "How
Much Has Output Declined in Eastem Europe
and the Former Soviet Union?," World Bank
Mimeo
Pohl, Gerhard and Piritta Sorsa. 1992.
European Integration and Trade with the
Developing World. Policy and Research Series
21. Washington D.C.: The World Bank.
Rollo, Jim and Alasdair Smith. 1993. "The
Political Economy of Eastern European Trade
with the European Community: Why So
Sensitive?," Economic Policy, 16:140-8 1.
Winters, L. Alan. 1992. "The Europe
Agreements: With a Little Help from Our
Friends," in The Association Process: Making it
Work. London: Center for Economic and Policy
Research.



17
Notes
1.    Reported GDP numbers and Growth rates are highly unreliable. Simple quantitative
indicators, such as electric power consumption changes may be for more reliable (see Pohl and
Dobozi, 1995).
2.    While official GDP estimates indicate a cumulative decline in economic activity of 21 %
during 1989-94 for both the Czech Republic and Hungary, electric power consumption in the Czech
Republic declined by only 11% but in Hungary by 18% during the same period. This is probably a
more consistent estimate.
3.    See, for example, Messerlin (1993); Pohl and Sorsa (1992), and Winters (1992).
4.    The data used are the most detailed, disaggregated trade figures that are available from the
statistical office of the European Communities (EUROSTAT) on a comparable basis: 8-digit
Combined Nomenclature (CN), the EU's version of the Harmonized System, or about 9,750 items.
The COMEXT CD-ROM database was used. It should be noted that the use of EU import data to
infer CEEC exports will inflate the level of exports as the EU values imports on a c.i.f. basis, i.e.,
inclusive of freight and insurance. It should also be noted that there are numerous instances where
reported imports by the EU are zero in 1989, but are quite large in 1993. To take just one example,
exports of copper and silver from Bulgaria to the EU was reported to be zero in 1989, and to have
grown to more than ECU 20 million in 1993. In the discussion that follows this will be interpreted as
a reorientation of trade. However, at this level of disaggregation it is quite possible that values for
specific items are zero for one or more years and that there may be inaccuracies in recording trade
flows.
5.    The main exception is Bulgaria. Thus, while Bulgaria has been able to reorient itself towards
the EU-it was heavily dependent on the USSR-this has taken place in the context of a severe fall in
output. Total trade is still below the level of the late 1980s.
6.    Aggregate data for the Czech and Slovak Republics are used in most of what follows in order
to maintain comparability. The same reasoning applies to the FSU, with the added rationale that this
allows us to avoid dealing with issues of transshipment and arbitrage activities. Separate trade data
for the Czech Republic and Slovakia are available for 1993 only. They reveal that the Czech
Republic now accounts for 80 percent of the combined exports. The composition of exports is also
similar, with the following exceptions: Slovakia is much more specialized in clothing and steel
production (14 and 19 percent of total exports in 1993, as compared to 8.5 and 12 percent for the
Czech Republic), while the Czech's are more specialized in machinery and transport equipment (18.5
and 9 percent, versus 10 and 6 percent for Slovakia).
7.    It should be noted that there are valuation problems here, both widh respect to absolute value
of exports in 1989 and as regards year-to-year changes (the latter due to variations in the ECU/dollar
exchange rate). The data for total exports in 1989 for Bulgaria, the Czech and Slovak Republics and
Romania have been adjusted for the overvaluation that results if official 'exchange rates' for
coiivertible rubles are used. It should also be kept in mind that by using EU import data, implied
exports .rom the CEECs are overvalued by the inclusion of freight and insurance.



18
8.    The CSFR had the greatest dependence on the CMEA: some 80 percent of its exports went to
centrally-planned economies, as opposed to 50 percent for Hungary, and 40 percent for Poland
(Halpern, 1994). It should be noted that as oi the end of 1990, the German Democratic Republic's
external trade is included in that of Germany in the EU's database. There is consequently a one-time
upward shift in the EU's imports from the CEECs.
9.    These figures are in constant 1988 dollars to ensure comparability with those reported by
Mastropasqua and Rolli (1994).
10.   The sample includes the CEECs. the FSU, Turkey, China, South Korea, North Africa, and
the Middle East. The change in the structure of exports is defmed as:
ja,,9-a,.31
E'.1 aiDg
where a; is the value of exports of an 8-digit CN item in 1989 and 1993, respectively, and K is the
number of commodities exported.
11.   The structural change in the FSU accelerated significantly in 1993. See Appendix Table 2. It
can be observed that on a year-to-year basis the largest values of changes in exports appear to occur
in years where price liberalization occurred, i.e., 1990 for Hungary and Poland; 1991 for the Czech
and Slovak Republics; and 1993 for imuch of the FSU.
12.   In the case of 'expired' commodities, the denominator used is total exports of the previous
year; in the case of new products, the denominator is total exports for the current year. Data can be
found in Appendix Table 3.
13.   Relative specialization is defined as:
zijixi
1xI1 xi/   X
where x, are exports of commodity i by countryj, XJ are country j's total exports, and N is the
number of countries. Usually this indicator, often called the revealed comparative advantage (RCA)
index, is defined for the world, i.e., the denominator would sum across all countries. In the case of
this paper however, specialization indices have been defined relative to total EU imports. Thus, the
denominator consists of total EU imports of commodity i, divided by total EU imports of all
commodities. Intra-EU trade is excluded because intra-EU trade is not available for 1993.
14.   Under central planning, Bulgaria became increasingly specialized in heavy industries such as
steel, and machinery and equipment. With the decline of Eastern narkt, these products could not
be sold on Western markets (Borensztein et al. 1993). The shift towards lighter industry revealed by
the export data therefore represents restructuring.



19
15.   Intra-industry trade is often calculated using the Grubel-Lloyd measure, which is defined as:
1 -EIX,-M,I
r t  (xi +M,)
where X, and Ml are a country's exports to--and imports from-a trading partner of commodity i,
respectively. The closer to one, the higher is intra-industry trade.
16.   See Greenaway and Hine (1991) for a recent survey of the theory and evidence in the EU
context. This is not to say that intra-industry trade will not lead to adjustment and thus pressure for
protection. To the extent that there are specific and relatively immobile factors of production that are
injured by import competition, they can be expected to seek protection. But the factors that are hurt
will be at the firm-level. Other firms in the industry will expand. This malces it more difficult to
obtain protection, as there will be coxi.iicting interests within industries.
17.   Processing trade has even expanded to agricultural goods. Almost 5 percent oF Poland's
agricultural exports to the EU enter under the outward processing regime. Apparently this is mostly
due to the German fish industry shipping raw crustaceans to Poland for shelling be hand (Naujoks and
Schmidt, 1994).
18.   Subcontracting activities are important here. Their share in total exports increased from 4 to
19 percent, and 2 to 13 percent for machinery and instruments, respectively, in the 1989-93 period.
19.   Greater than ECU 10 million in 1989.
20.   Note that these calculations assume that the zero imports reported by the EU for a number of
the goods in question reflect reality.
21.   See Rollo and Smith (1993) and Kaminski (1994) for more detailed analyses.
22.   See Neven (1994) for a detailed and careful analysis of the likely political economy of
pressures for contingent protection in the EU against the CEECs. Romania has a lot of competitor
countries that have a somewhat similar pattern of exports to the EU. These countries include the
Philippines, China, Turkey, Hong Kong, and Taiwan. Presumably this reflects the great increase in
relative specialization in textiles and clothing in Romania. In the case of Hungary Mexico and Brazil
have become 'competitors', while the Czech and Slovak Republics have become more similar to
Turkey and Taiwan. Poland's pattern of relative specialization does not appear to be similar to any of
the countries chosen as potential comparators, with the exception of the FSU. The 1993 correlation
coefficient between Poland-FSU is 0. 19, up from zero in 1989.
23.   Any rationale for the maintenance of antidumping in EU-CEEC trade disappears once the
CEECs have abolished trade barriers and adopted the EU's competition policies. As argued at greater
length in Hoekman and Mavroidis (1994), efforts should be made to link the elimination of
antidumping to the full implementation of the Europe Agreements.



Appendix Table I
Exports to the EU, 1989 and 1993
(ECU million and percentage)
Value      Value      Market     Market      Average    Export/    Predicted    EU Share  Predicted
Country / Group     1989        1993      Share in    Share in   Annual       GDP    Export/GDP*  in Total   EU Share
EU         EU       Change in     1993                   Exports   of Total
1989       1993      Exports to                                    Exports*
EU
1989-93
Bulgaria               523         944         0.12        0.19        15.9      35.1          21        30.4       44.6
Czech & Slovak Rep.    2,521     5,981         0.56        1.23       24.1       31.9          28        52.4       65.0
Hungary              2,571       3,931         0.58        0.81        11.1      23.4          31        51.8       64.0
Poland               3,817       7,505         0.85        1.54        18.4      16.2          19        62.2       55.0
Romania              2,531       1,670         0.57        0.34       -9.9       19.7          23        40.0       48.0
FSU                  13,258     15,457         2.97        3.18        3.9        10.6          9        32.4       50.0
China                 9,077     19,491         2.03        4.01        21.1       17.7
South Korea          6,925       7,690          1.55        1.58       2.7       24.9
Turkey                5,509      6,530          1.23        1.34       4.3        13.0
Morocco              2,677       3,195         0.60        0.66        4.5        13.9
CEECs                11,964     20,030         2.68        4.12        13.7
MENA                26,595      29,537         5.95        6.08        2.7
*      lHamilton and Winters (1992).
Baldwin (1994).



Appendix Table 2
Total Change in Structure of Exports
(8 digit level)
Country ! Group      1989 -    1990 -     1991 -    1992-    Change   % of change l
1990      1991       1992      1993      89- 93      due to
l_________                     __________    grow th
Bulgaria                  1.49       1.71       1.26      1.10       1.53         71.5
Czech and Slovak Rep.     0.85       1.60       1.03     0.76        1.67         83.6
Hungary                   1.04       1.05       0.79     0.81        1.14         68.9
Poland                    1.35       1.01       0.76     0.79        1.38         78.7
Romania                   1.11       0.86       0.93     0.89        0.71         31.9
FSU                       0.60       0.67       0.81      1.33       0.72         59.4
China                     0.91        1.09      0.55     0.58        1.30
South Korea               0.95       0.82       0.86     0.80        0.91:
Turkey                    0.93       0.73       0.64     0.52        0.77
Morocco                   0.83       0.87       0.64     0.65        0.80
Note:  See iext for the definition of this variable. All variables are expressed relative the annual mean.



Appendix Table 3
Relative Impoilance of Expired and New Commodities
(percent)
1990           1991           1992            1993          1989-93
Bulgaria                        6.91          10.65            9.33            9.48           23.7
12.86          12.56            8.26           9.96            31.8
Czech & Slovak Rep.             3.45           3.27            2.99            3.48           11.5
4.18           7.40            2.48           5.37            19.8
Hungary                         4.16           2.44            2.68            8.09           18.6
5.10           3.41            3.11           8.27            20.7
Poland                          4.46           1.76             1.65          4.62            13.8
6.28           2.37            2.28           5.27            15.7
Romania                         5.46           3.63            5.53            6.17           23.9
4.15           5.35            5.79           7.47            18.0
FSU                             1.38           0.96            8.46            1.89            7.9
0.86           2.94            1.29           3.72             9.5
China                           1.49           0.66            2.54            4.04            7.9
1.79           1.05            2.67           3.66            10.8
South Korea                     5.52           2.38            7.89            6.48           17.3
6.94           1.9i            7.82           6.90            18.4
Turkey                          3.62            1.93            3.23           5.29            12.7
3.84           1.71            2.20           4.65            10.7
Morocco                         2.56           8.31            3.72           11.82            16.7
2.32           7.99            3.61           5.42            15.8
Note:  The first line for each country is the share of exports in the previou.s year that disappeared; the second line for each country is the share of exports in the current
year that is accounted for by 'new' products. The first row of the 1989-93 column is relative to 1989; the second row is relative to 1993 total exports.



Appendix Table 4: Share of 'Traditional' Products in Total Exports, 1989-93
(ECU million and percentage)
8-Digit Products > ECU 3 million in 1989   8 Digit Products < ECU 3 million in 1989
1989          1993          Annual         1989           1993          Annaa
Average                                    Ave
Change                                     Change
Bulgaria               230            182           -5.6           293            761           27.0
-___________          (44.0)         (19.3)                       (56.0)         (80.7)
CSR                   1,519          1,784           4.1          1,002          4,197          43.1
--- _________         (60.3)         (29.8)                       (39.7)         (70.2)
Hungary               1,513          1,057                        1,390         2,541           24.5
(58.9)        (35.4)          -2.1         (41.1)         (64.6)
Paland                2,843           974            7.2          3,748          3,757          40.1
(74.5)        (49.9)                       (28.5)         (50.1)
Romlana               2,036           860                          496            810           13.1
(80.4)        (51.5)          -19.4        (19.6)         (48.5)
FSU                  12,704         12,295                         553           3,162          54.6
,__________   ~ (95.8)            ( (79.5)        -0.8         (4.4)          (20.5)
Figures in parenthaes are shams in total exports.



Appendix Table 5: Relative Speciallization of Exports to the EU, 1989 and 1993
Agric.   Ores/   Leather  Wood   Fibers  Clothing    Clay/   Iron/  Nonter.  AMachines  Transp.  Instru-  Arms  Fumi-
Chem.                 _                    Glass   Steel   Metals             Equip.  ments           ture
Bulgaria    2.51    0.93     0.83    0.67    0.90     2.01      0.43   3.96     0.56     0.55     0.10    0.17   1.28    2.24
1.95    0.75     3.50    0.56    1.66     3.01      1.04   1.35     3.02      0.45     0.07    0.18   0.16    1.23
CZSL       0.94     0.95     1.24    1.80    1.39     1.40      1.72   4.98     0.11      0.41    0.89    0.28   7.62    3.61
0.51    0.72      1.70    1.08    1.65     1.51     1.19   5.40     0.94      0.66     1.09    0.29   1.85    4.60
Hungary    3.20     0.63     2.39    0.53    0.89     2.80      0.53   2.43     1.12      0.54    0.15    0.15   1.29    3.25
2.11    0.55     2.60    0.55    0.73     2.54      0.51   2.30     1.43      0.87     0.40    0.23   0.64    2.92
Poland     2.55     0.92     1.21    0.72    0.37     2.11      0.59   3.01     2.02     0.35     0.67    0.11   0.56    4.70
1.26    0.73     1.19    1.17    0.55     2.70      0.45   3.21     2.30      0.35     1.33    0.12   0.08    6.25
Romania   0.43      1.61     0.91    0.56    0.61     3.57      0.54   2.12     1.56      0.20    0.24    0.05   0.02   17.82
0.i6    0.39     4.20    0.41    0.77     5.77      0.56   2.76     0.29      0.29     0.31    0.10   0.44   13.07
FSU        0.18     2.57     0.52    1.09    0.85     0.04      1.31   1.21     2.63      0.05    0.36    0.07   1.12    0.27
0.36    2.48     0.68    0.74    1.88     0.27       1.14   1.68    5.54      0.05     0.24    0.06   0.49    0.43
China      1.15     0.53     4.49    0.37    3.69     4.04      0.44   0.62     0.22      1.21    0.08    0.62   0.35    1.36
0.69    0.37     5.22    0.37    1.48     3.30      0.42   0.67     0.31      1.43     0.08    0.83   0.22    2.21
South      0.16     0.20     7.30    0.10    1.51     3.38      0.39   0.75     0.07      2.04    0.54    0.79   0.10    0.28
Korea      0.15     0.38     3.89    0.11    1.68     1.38      0.25   1.00     0.15      1.83     1.76    0.79   0.33    0.11
1.57    0.64     3.35    0.03    4.41     7.43 '    0.61   1.70     0.48      0.20     0.21    0.07   1.15    0.48
Turkey     1.85     0.35     2.45    0.07    3.38     7.03      0.59   0.72     0.43      0.29    0.33    0.07   2.05    0.55
2.93    0.96      1.97    0.32    0.66    7.02      0.28   0.16     0.40      0.19     0.34    0.05   0.01    0.29
Morocco    2.76     0.68     1.67    0.19    0.64     6.80      0.17   0.18     0.36      0.28    0.09    0.08   0.13    0.15
2.43     1.04    0.24    0.16    1.66      1.91     3.88   0.19     0.42      0.58     0.18    1.03   4.49    0.77
Israel     1.98     1.09     0.18    0.18    1.42     1.66      2.15   0.16     0.74      0.82    0.13    1.19   0.54    0.78
0.38    2.80     0.06    0.01    3.89     0.49      0.11   0.72     2.32      0.04     0.28    0.06   0.02    0.07
Egypt      0.61     2.89     0.17    0.03    4.20     1.06      0.12   0.41     1.46      0.08    0.52    0.14   0.14    0.23



Appendix Table 6
Relative Specialization: Correlation Coefficients, 1989 and 1993
1993
_____________ ___,.,,__, , BUL         CSR        HUN         POL       ROM
Bulgaria                              1.000
Czech and Slovak Rep.                 0.207      1.000
Hungary                              0.349       0.277      1.000
Poland                               0.283       0.516      0.597       1.000
Romania                              0.244       0.450      0.327       0.490     1.000
FSU                                  0.202                              0.190
Philippines                                                                      0.352
China                                                                            0.248
Turkey                                           0.238      0.080                0.101
Brazil                                                      0.104
Hong Kong                                                                        0.161
Thailand                                                    0.163
Taiwan                                           0.184                           0.166
Mexico                                0.426                 0.213
Morocco                               0.092 1         I           I                    1
1989
BUL        CSR        HUN         POL       ROM
Bulgaria                              1.000
Czedh and Slovak Rep.                 0.151      1.000
Hungary                               0.418      0.423      1.000
Poland                                0.484      0.161      0.616       1.000
Romania                               0.133      0.174      0.202       0.168     1.000
FSU                                   0.079                             0.003
Philippines                                                                      0.315
China                                                                             0.140
Turkey                                           0.340      0.170                0.045
Brazil                                                      0.021
Hong Kong                                                                        0.100
Thailand                                                    0.183
Taiwan                                          -0.012                           0.119
Mexico                                0.228                 0.070
Morocco                               0.094



Policy Research Working Paper Series
Contact
Title                           Author                  Date             tor paper
WPS1411 Income Inequality, Welfare, and  Nanak Kakwani          January 1995       G. Evans
Poverty: An Illustration Using                                            85783
Ukrainian Data
WPS1412 Foreign Technology Imports and   Xiaorming Zhang        January 1995       C. Jones
Economic Growth in Developing   Heng-lu Zou                               3775-
Countries
WPS1413 Endogenous Distortions in Product   Martin Rarna        January 1995       S. Fallon
and Labor Markets               Guido Tabellini                           38009
WPS1414 The World Bank and Legal Technical The World Bank       January 1995       K. Mathemovi
Assistance: Initial Lessons     Legal Departmnent                         82782
WPS1415 China's GDP in U.S. Dollars Based    Ren Ruoen          January 1995       E. O'Rielly-Campbell
on Purchasing Power Parity      Chen Kai                                  33707
WPS1416 Informal Regulation of Industrial  Sheoli Pargal        February 1995      E. Schaper
Pollution in Developing Countries:    David Wheeler                       33457
Evidence from Indonesia
WPS1417 Uncertainty and Global Warming: An  Andrea Baranzini    February 1995      C. Dell
Option-Pricing Approach to Policy    Marc Chesney                         85148
Jacques Morisset
WPS1418 The Impact of Labor Market       Lyn Squire             February 1995      G. Bayard
Regulations                     Sethaput Suthiwart-                       37460
Narueput
WVPS1419 Industry Structure and Regulation    Martin C. Stewart-Smith   February 1995  N. James
82758
WPS1420 Legislative Frameworks Used to   William T. Onorato     February 1995      W. Onorato
Foster Petroleum Development                                              81611
WPS1421 Distribution of Income and the Income Zeliko Bogetic    February 1995      F. Smith
Tax Burden in Bulgaria          Fareed M. A. Hassan                       36072
WPS1422 Efficiency in Bulgaria's Schools:  Zeljko Bogetic       February 1995      F. Smith
A Nonparametric Study           Sajal Chattophadyay                       36072
WPS1423 The Role of Commercial Banks in  Millard Long           February 1995      R. Gamer
Enterprise Restructuring in Central  Izabela Rutkowska                    37670
and Eastern Europe
WPS1424 Terms-of-Trade Shocks and Optimal  Luis Serven          February 1995      E. Khine
Investment, Another Look at the                                           37471
Laursen-Metzler Effect



Policy Research Working Paper Series
Contact
Tite                            Author                  Date              for paper
WPS1425 On the Intersectoral Migration of  Donald Larson         February 1995      J. Jacobson
Agricultural Labor              Yair Mundlak                              33710
WPS1426 Russian Unemployment: Its         Simnon Commander       February 1995      V. Reid
Magnitude, Characteristics, and  Ruslan Yemtsov                           35195
Regional Dimensions
WPS1427 Corporate Govemance and Equity   Stijn Claessens         February 1995      F. Hatab
Prices: Evidence from the Czech                                           35835
and Slovak Republics
WPS1428 Short-Term Supply Response to a    Bruno Boccara         February 1995      M. Pfeiffenberger
Devaluation: A Model's Implications  Fabien Nsengiyumva                   34963
for Primary Commodity-Exporting
Developing Countries
WPS1429 The World Trade Organization's    Bemard M. Hoekrnan     March 1995         F. Hatab
Agreement on Govemment          Petros C. Mavroidis                       38535
Procurement Expanding Disciplines,
Declining Membership?
WPS1430 Intergovemmental Fiscal Relations   Richard M. Bird      March 1995         G. Coward
and Poverty Alleviaton in Viet Nam   Jennie I. Litvack                     80494
M. Govinda Rao
WPS1431 The Industrial Pollution Projection    Hernamala Hetfige  March 1995        A. Williams
System                          Paul Martin                               37176
Manjula Singh
David Wheeler
WPS1432 Using Financial Futures in Trading    Ignacio Mas        March 1995         K Binkley
and Risk Management             Jesbs Saa-Requejo                          81143
WPS1433 Enterprise Restructuring in Eastem   Bemard Hoekman      March 1995         F. Hatab
Europe: How Much? How Fast7     Gerhard Pohl                              35835
Where? Preliminary Evidence from
Trade Data