The "Soaring Eagle": Anatomy of the Polish Take-Off in the 1990s.

De Broeck, Mark ; Koen, Vincent

I. Introduction

Poland's growth record through 1998 stands out on two counts. Following a sizable contraction around the turn of the decade, from 1992 Poland enjoyed seven years of uninterrupted growth at an annual rate averaging over 5 percent. This expansion has been remarkably vigorous and resilient compared with performance during the 1980s, outstripping even the more optimistic forecasts. It has also been far more impressive than what has been observed in other countries in transition, and on average over twice as rapid as in EU countries or as in OECD countries at large. This economic revival has earned Poland some flattering metaphors, in particular those of "East-European Tiger" and "Soaring Eagle".

This paper documents the key dimensions of Poland's growth performance in the 1990s. Rather than introducing a new formal framework of analysis or engaging in econometric testing, it draws on a wide range of relevant statistics and along the way discusses to what extent they square with common stylized views of transition. The paper first puts the country's overall growth performance in perspective, asking how remarkable it really is. It then turns to the sectoral, ownership, and regional patterns of growth, with a view to discovering how broad-based the expansion has been; particular attention is devoted to the performance of industry, which accounts for a large chunk of total value-added. The dynamics of the recovery and take-off are then discussed, to determine to what extent they reflected improved resource use versus factor accumulation. The concluding sections look ahead by discussing the prospects for medium-term growth and convergence towards EU per capita income levels.

II. Poland's Growth Performance in Perspective

The stabilization and liberalization of the economy initiated in 1989 was associated with a sizable contraction (Balcerowicz, 1995). While the contraction was deeper than any earlier post-World War II recession in Poland, it was shallower and shorter than in most other transition countries. The recovery started around late 1991, with growth gradually spreading from industry to other sectors and remaining at or above 5 percent per annum between 1994 and 1998, while investment boomed. During 1992-98, real GDP increased at an average rate of 5.2 percent per annum, and over the five years to 1998, the annual growth rate averaged 6.0 percent. In cumulative terms, the Polish economy thus expanded by 42 percent between 1991 and 1998 (and by over one third between 1993 and 1998).(2) By several measures, this performance is enviably good.

Firstly, these outcomes were consistently better than anticipated. At the onset of the transition in 1989, Poland was considered to be in a deep crisis and to be facing more difficult economic challenges than such neighboring countries as Czechoslovakia and Hungary. Ill-conceived partial reforms undertaken in the final years of the old regime had resulted in major macroeconomic imbalances, weak corporate governance and a lack of wage discipline, and asset stripping by insiders.(3) From 1993 and except in 1998, growth systematically and significantly exceeded the official one-year-ahead projections embedded in the budget laws. Likewise, growth turned out to be stronger than projected in the 1994-97 Strategy for Poland medium-term framework of the government, which ex ante was generally considered as rather optimistic. Growth was also distinctly higher than what most observers foresaw: for instance, in what they described as their optimistic long-run scenario, Czyzewski et al. (1994) showed an average annual growth rate of 4.5 percent for the period until 2005; subsequent IMF simulations also pointed to a medium-term equilibrium growth potential on the order of 4 [degrees] to 5 percent per annum (IMF, 1996).

Secondly, this growth spell contrasts with the lackluster record of the previous fifteen years.(4) Since the early 1970s, Poland had not witnessed this long and forceful an expansion and had been underperforming the other planned economies in central and eastern Europe In this context, the historical peak level of measured real GDP of 1989 was already exceeded by 1995 (Figure 1).(5) In fact, to the extent that even the revised official series may still understate GDP and that the composition of output improves under market-based rules, the recovery was actually even swifter.(6)

[GRAPH OMITTED]

Thirdly, Poland's growth record over the seven years to 1998 compares very favorably with that of its relatively successful neighbors in central and eastern Europe (Figure 1) and the Baltics, and a fortiori with developments in Romania, Bulgaria, and countries of the Commonwealth of Independent States (De Broeck and Koen, 2000). It is also more impressive than Slovenia.

Poland's earlier, more sustained, better balanced and stronger growth was related to a combination of relatively favorable initial conditions and external factors. Poland had made significant progress toward price and trade liberalization during the 1980s. At the outset of the transition, the share of private sector activities was sizable, in particular in agriculture. The quality of the country's public sector institutions and legal system was relatively high, and indicators of educational attainment and the quality of schooling suggested high levels of human capital. In combination with a relatively low per capita income level, these favorable initial conditions resulted in a considerable growth catch-up potential. Moreover, Poland enjoyed a close proximity to the European Union (EU) countries and unlike some other transition countries did not experience internal or external conflicts in the early transition years.

In addition to these initial conditions and factors, Poland's robust growth performance also reflects macroeconomic and reform policies that, on the whole, were sound. More specifically, the policy ingredients of the Polish success story include:(7)

* an early political window of opportunity opening in 1989 (the so-called period of "extraordinary politics," characterized by a broad political consensus on the reform strategy and a willingness among the public to accept difficult and radical measures);

* a comprehensive and upfront further price and trade liberalization, extending the initial liberalization measures introduced in the 1980s;

* current account convertibility of the currency and an early and broad dismantling of obstacles to foreign trade, which prompted a swift reorientation of trade toward the West and put pressure on firms to restructure;

* low entry barriers for new firms, facilitating a redistribution of labor from state-owned to new enterprises;

* the imposition of relatively hard budget constraints on public enterprises;

* a relatively liberal social safety net -- a generous retirement regime during the early transition years in particular -- easing the social strains associated with restructuring;

* a consistently prudent macroeconomic policy stance, including an exchange rate policy geared to avoid overvaluation;

* a substantial reduction in external debt, which opened the door to large-scale foreign direct investment (FDI) inflows;

* the establishment of a commercial banking sector and securities markets subject to strict regulation and strong supervision, and the implementation of a financial restructuring program providing incentives to banks and enterprises to resolve non-performing loans;

* extensive legal and institutional reform in support of the market and private sector activity in non-financial areas as well.

The remainder of the paper focuses on three distinct features of the Polish experience that can largely explain the country's strong growth record. First, rapid and comprehensive price and trade liberalization and the imposition of financial discipline on economic agents, state-owned enterprises in particular, resulted in a swift and massive reallocation of resources toward market-determined uses. Second, strong expansion in newly established domestic small- and medium-sized enterprises (SMEs) -- facilitated by liberal regulations affecting entry, fast small-scale privatization, and the release of labor and capital by financially constrained state-owned enterprises -- acted as a main source of growth in the initial transition years. Third, factor accumulation, especially as regards capital, became the dominant source of growth from 1994. The increase in investment was driven in part by a boom in FDI, and foreign affiliates replaced local SMEs as providers of the main impetus to growth.

While Poland's performance clearly stands out vis-a-vis the contemporaneous experience of other transition countries, it has been surpassed in other regions over the 1992-98 period. Among European countries, Ireland has enjoyed far more rapid growth over this period (averaging 7 [degrees] percent per annum). More distant "tigers," particularly in Asia, have recorded even faster growth during those years. In fact, most countries in the world have experienced at least one 7-year expansion at 5 percent per annum or higher since the 1970s, including over half of the current OECD member countries. Admittedly, however, few among those countries had to cope with as momentous a structural transformation as the transition countries had in the 1990s. Moreover, most of those episodes of rapid growth had a broad regional basis, while Poland's performance did not.

Poland's relative position in Europe has improved considerably during the 1990s (Figure 2). In terms of absolute size gauged at market exchange rates, Poland caught up with the poorer EU member countries (namely, Greece and Portugal). Poland's economic weight also increased compared with its immediate neighbors: while Poland's GDP was 7 percent larger than the combined GDP of Hungary, the Czech Republic and the Slovak Republic in 1991 (at market exchange rates), it had become 21 percent larger by 1998.(8) In per capita terms, and at purchasing power parity rates, Poland's relative position improved sharply, in contrast to that of its aforementioned neighbors. Thus, per capita income rose from 46 to 54 percent of the level in Greece.(9) Notwithstanding the dynamism of the catch-up, however, living standards in Poland remain well below those in Hungary, the Czech Republic and even the Slovak Republic.

[GRAPH OMITTED]

III. Sectoral, Ownership, and Regional Variations

Disaggregating value-added by sector, ownership category, and region allows us to identify the main sources of growth. Decomposing the sectoral and regional pattern of growth also gives an indication of how widespread the overall expansion has been, bearing in mind that the more diversified the growth base, the less vulnerable growth is to sector or region-specific shocks (although, at the same time, large shifts in the composition of output at finer disaggregation levels can be indicative of restructuring). Despite the relatively poor quality of the available data, the changes over the period under consideration are large enough to allow some broad indicative conclusions.

Growth has been driven first and foremost by industry, and more specifically by manufacturing (Figure 3 and Table 1). More than 60 percent of the cumulative increase in aggregate value-added between 1991 and 1998 is accounted for by manufacturing. Over that same period, mining and agriculture essentially stagnated, notwithstanding some short-run fluctuations. Construction contributed a bit more than proportionately, largely reflecting strong growth of this sector in 1997-98. Somewhat surprisingly, market services,(10) at least as measured in the national accounts, started to grow later and less vigorously, and in cumulative terms contributed much less than proportionately. At the broadest level of aggregation, there has thus been less of a shift from industry to services than might have been expected on the basis of the widespread perception of the former regime as being characterized by overindustrialization. The decline of the relative importance of the loss-making mining sector, however, is more in line with what could have been anticipated, as is the diminishing weight of the agricultural sector, which at the onset of the transition was larger than in comparable transition economies (Landesmann, 2000). Also, the broadly defined services sector comprises both activities that were overweight under central planning -- freight and certain government administrative functions, for instance -- and activities that were underweight, including trade, hotels and restaurants, and financial intermediation. A strong expansion in the provision of such consumer-oriented services since 1992 provided a major impetus to overall growth, as had been expected. Moreover, this expansion is likely to be underestimated in the official statistics as tax-induced under-reporting affects consumer-oriented services in particular.

[GRAPH OMITTED]

Table 1:
Growth of Value Added by Sector
(In percent, in real terms)

 Sectoral
 share
 in
 1992 1992 1993 1994

Agriculture, forestry 6.7 -9.4 7.0 -15.1
Fishing 0.1 ... -53.9 10.6
Industry 34.0 2.6 8.6 10.3
of which:
 Mining and 3.4 -5.3 -9.8 -1.9
 quarrying
 Manufacturing 26.9 5.1 11.9 11.2
 Electricity, gas 3.8 ... 1.5 16.2
 and water supply
Construction 7.8 3.8 1.1 2.7
Trade and repair of 13.1 ... 5.8 -1.5
 consumer goods
Hotels and restaurants 0.4 ... 2.3 7.3
Transport, storage 6.2 ... -5.3 0.5
 and communication
Financial intermediation 0.5 ... 29.1 102.1
Real estate and business 6.5 ... 1.5 6.9
 services
Public administration 6.1 ... 5.2 7.8
 and defense
Education 3.8 ... 0.1 10.3
Health care and 4.2 ... 0.5 4.1
 social security
Other 6.5 ... -18.2 3.4

Memo: GDP 2.6 3.8 5.2

 1995 1996 1997 1998

Agriculture, forestry 10.4 2.4 1.1 6.1
Fishing -8.3 7.0 -10.5 -33.0
Industry 10.4 7.6 10.3 4.3
of which:
 Mining and 1.5 4.7 -4.3 -11.6
 quarrying
 Manufacturing 13.7 8.8 14.4 7.6
 Electricity, gas 2.1 3.5 0.6 -1.0
 and water supply
Construction 5.8 2.8 13.6 9.1
Trade and repair of 5.0 6.1 8.1 5.1
 consumer goods
Hotels and restaurants 6.5 16.8 7.3 8.4
Transport, storage 2.3 5.4 5.6 6.3
 and communication
Financial intermediation 21.4 11.3 2.9 6.6
Real estate and business 5.8 0.9 -0.7 4.0
 services
Public administration 4.2 4.4 5.9 1.6
 and defense
Education 1.5 1.4 0.9 2.7
Health care and 1.6 1.5 0.8 1.5
 social security
Other 0.7 6.9 -6.8 1.0

Memo: GDP 7.0 6.0 6.8 4.8

Source: GUS and authors' calculations.

(1.) Based on the 1993 system of national accounts starting from 1993. The 1992 figures are based on the former system and therefore incomplete and not directly comparable to subsequent ones.

Another feature of growth in the 1990s is the greater dynamism in the private sector. Between 1991 and 1998, the private sector share in output rose at a faster pace than the share in employment, at both the aggregate and the broad sectoral levels, in industry in particular.(11) In 1998, the private sector accounted for around 70 percent of aggregate output and employment. The observation that the private sector outperformed its state sector counterpart holds even taking into account the fact that -- to an extent that official statistics do not allow to identify -- private sector growth in part reflects privatization operations (Kennedy, 1997). The expansion of the private sector was first driven by new domestic SMEs. From 1994, stimulated by an agreement on external debt owed to the London Club of creditors and by an upswing in large-scale privatization, FDI began to pick up, and the main driving force of private sector expansion changed to foreign affiliates. Reflecting the increasingly important contribution from these affiliates to growth, their share in manufacturing employment rose from less than 10 percent in 1993 to more than 20 percent in 1998, while that in manufacturing sales rose from somewhat more than 10 percent to more than 30 percent in the same period.

The increase in the share of private sector activity has been associated with a rapid expansion in the number of firms, starting from a highly concentrated pattern, in industry in particular. At the end of the 1980s, less than 3 percent of the state-owned 4,900 firms in industry accounted for half of industrial output, and large firms also accounted for the bulk of industrial employment (Berg, 1994). Between 1991 and 1998, the number of registered commercial law companies increased more than two and a half times, to 136,500, that of joint ventures rose almost eight-fold to around 37,000, while that of unincorporated private businesses increased by more than 50 percent, to more than 2 million.

The strong growth in the number of private sector enterprises reflects both a correction to the distorted size distribution of firms in industry and the development of a small business sector oriented toward consumer services (Table 2). This growth was supported by liberal entry requirements and by privatization policies that quickly transferred SMEs in the services sector to private ownership and that gave state enterprises in other sectors incentives to sell or lease assets to private enterprises. Moreover, new private firms benefited from a relatively easy access to bank lending, as the reform and reorganization of the banking sector contributed to a reorientation of credit flows to such firms (Anderson and Kegels, 1998; Bratkowski, Grosfeld, and Rostowski, 2000). The rapid development of a new private enterprise sector is one of the key features of the so-called Polish model of transition.(12)

Table 2:
Indicators of Private Sector Dynamism

 1989 1990 1991 1992 1993 1993(1)

Number of enterprises

Commercial 16905 36267 53771 69907 83283 83283
 law companies
of which with 429 1645 4796 10131 15167 15167
 foreign capital
of which in:
Industry 4083 9073 11735 15315 18338 19862
Construction 3406 5946 9632 12063 13775 11667
Trade 2175 9043 17768 24199 29381 30444
Hotels and ... ... ... ... ... 971
 restaurants
Transport and 1672 2081 3284
 communication
Financial ... ... ... ... ... 603
 intermediation
Unincorporated ... 1135 1420 1629 1862 1862
 private businesses
 (in thousands)
State enterprises 7337 8453 8228 7245 5924 5924
Memo item
Started 107 1297 2056 2635 2635
 privatizations
Completed 6 228 612 989 989
 privatizations

Private sector
share in output
(in percent)

Gross domestic ... 30.9 41.7 45.2 47.9 46.4
 product
Agriculture ... 77.5 80.0 84.6 86.5 86.5
Industry(2) 16.2 18.3 24.8 28.2 34.6 34.6
Mining ... ... ... ... ... ...
Manufacturing ... ... ... ... ... ...
Construction(2) 25.5 33.8 58.8 74.3 82.3 82.3
Retail sales 59.5 63.7 82.8 86.4 89.1 89.1
Transport and ... ... 35.1 37.3 37.3
 communication(2)
Exports ... 4.9 21.9 38.4 44.0 44.0
Imports ... 14.6 49.9 54.5 59.8 59.8

Private sector
share in employment
(in percent)

Total 46.2 48.9 54.1 56.0 58.9 58.9
Agriculture 84.9 86.5 89.2 91.4 93.7 94.6
Industry 29.1 31.2 35.8 40.5 42.7 40.7
Mining ... ... ... ... ... 1.9
Manufacturing ... ... ... ... ... 49.2
Construction 37.4 42.1 59.5 71.9 74.8 69.6
Trade and repair 72.7 82.2 88.3 90.7 93.2 92.7
Transport, and ... ... 20.1 21.0 27.6
 communication
Financial ... ... 32.0 27.6 33.1 33.9
 intermediation
Real estate ... ... ... ... ... 57.9

 1994 1995 1996 1997 1998

Number of enterprises

Commercial 95017 104922 115739 26465 136497
 law companies
of which with 19737 24086 28622 32942 36850
 foreign capital
of which in:
Industry 22726 25077 27379 29532 31426
Construction 12821 13634 14646 15589 16565
Trade 34639 38001 41991 45896 49195
Hotels and 1230 1483 1771 2045 2328
 restaurants
Transport and 3950 4454 5003 5550 6064
 communication
Financial 851 1118 1377 1706 2012
 intermediation
Unincorporated 1977 1957 2241 2541 2637
 private businesses
 (in thousands)
State enterprises 4955 4357 3847 3369 2906
Memo item
Started 3132 3582 3953 4178 4648
 privatizations
Completed 1380 1690 1991 2325 2569
 privatizations

Private sector
share in output
(in percent)

Gross domestic 45.5 52.8 54.5 58.7 60.9
 product
Agriculture 87.9 89.1 88.7 88.7 89.5
Industry(2) 39.4 46.9 52.4 63.5 69.1
Mining 2.3 2.6 2.6 5.5 7.6
Manufacturing 47.0 53.5 62.1 75.2 80.6
Construction(2) 84.9 87.6 87.9 93.2 93.9
Retail sales 90.8 92.4 92.9 94.7 95.0
Transport and 38.5 40.1 42.2 44.7 ...
 communication(2)
Exports 53.2 56.8 62.9 74.3 78.8
Imports 66.9 69.7 75.6 82.5 86.5

Private sector
share in employment
(in percent)

Total 60.6 62.4 65.1 68.2 70.7
Agriculture 95.9 96.6 97.6 97.8 98.0
Industry 46.1 50.5 55.2 63.7 70.1
Mining 3.0 3.1 3.3 4.5 6.9
Manufacturing 55.1 60.0 64.9 74.5 81.5
Construction 78.2 80.9 84.5 87.7 91.3
Trade and repair 93.4 94.1 94.9 96.1 97.3
Transport, and 25.4 26.7 28.8 32.9 34.9
 communication
Financial 33.3 36.4 38.6 42.0 49.9
 intermediation
Real estate 60.0 63.3 65.4 69.8 73.3

Source: GUS and authors' calculations

(1) New classification;

(2) Share in sales

Turning to the regional dimension,(13) growth appears not to have been uniform. Homogeneous regional GDP data are not available throughout the period under consideration, and those that exist suffer from a major structural break in 1992.(14) Notwithstanding those limitations, it can be noted that the recovery has been pulled by the capital (Warsaw) and a small number of other large regions which have also recorded above-average growth. The Warszawkie (Warsaw) region alone accounted for around one fifth of nationwide growth between 1992 and 1996. The two largest regions (i.e., the capital and Katowice) accounted for over a third of aggregate growth. Close to 60 percent of growth was accounted for by only six of the 49 regions, representing no more than 36 percent of nationwide GDP in 1992.

Although regional and sectoral specificities overlap to some extent, regions burdened with stagnating sectors do not appear to be systematically lagging. For example, there is only an extremely weak correlation between the share of agriculture in regional GDP in 1992 and regional GDP growth since, and the main mining region (Katowice) has recorded above-average GDP growth. More detailed data indicate that both overall and industrial specialization in the regions declined markedly during the period of high growth 1994 to 1998 (Deichman and Henderson, 2000). This suggests that the local drag from struggling dominant sectors has often been compensated by the development of activity in others, which might constitute indirect evidence of restructuring.

Nonetheless, the uneven speed of the recovery across regions begs the question whether in the process regional per capita GDP has tended to become more or less equal. The top panel in Figure 4 shows that not all regions were equally affected during the great contraction of the late 1980s and early 1990s. The capital in particular was relatively spared, even though it too suffered a large decline. The bottom panel illustrates the unevenness of the recovery through 1996.(15) Table 3 displays a set of summary statistics pointing to a significant increase in regional disparities between 1992 and 1996, and also, albeit less unequivocally, between 1986 and 1996.(16) Cross-regional correlations highlight the magnitude of the shifts that took place, with regional rankings changing both during the second half of the 1980s and during the first half of the 1990s. (17)

[GRAPH OMITTED]

Table 3:
Regional GDPs: Selected Summary Indicators
(Summary statistics based on nominal regional GDP per
capita relative to the nationwide average)

 Including Warsaw

 1986 1990 1991 1992 1995 1996

Coefficient 0.32 0.36 0.34 0.23 0.32 0.34
 of variation
Maximum 2.15 2.35 2.27 1.58 2.10 2.13
Minimum 0.60 0.52 0.54 0.65 0.60 0.59
Max/min 3.6 4.6 4.2 2.5 3.5 3.6
Average of 2.19 2.53 2.35 2.03 2.22 2.30
 top 5 over average
 of bottom 5
Average of 1.83 2.08 2.01 1.80 1.89 1.91
 top 10 over average
 of bottom 10

Correlation matrix

 Including Warsaw

 1986 1990 1991 1992 1995 1996

1986 1.00 0.94 0.93 0.79 0.86 0.84
1990 1.00 0.97 0.86 0.90 0.89
1991 1.00 0.85 0.91 0.90
1992 1.00 0.87 0.87
1995 1.00 0.98
1996 1.00

 Excluding Warsaw

 1986 1990 1991 1992 1995 1996

Coefficient 0.24 0.26 0.26 0.21 0.29 0.30
 of variation
Maximum 1.55 1.67 1.61 1.49 2.10 2.13
Minimum 0.60 0.52 0.54 0.65 0.60 0.59
Max/min 2.6 3.2 3.0 2.3 3.5 3.6
Average of 1.79 2.00 1.88 1.92 1.95 1.97
 top 5 over average
 of bottom 5
Average of 1.64 1.84 1.79 1.73 1.75 1.77
 top 10 over average
 of bottom 10

Correlation matrix

 Excluding Warsaw

 1986 1990 1991 1992 1995 1996

1986 1.00 0.88 0.86 0.73 0.83 0.78
1990 1.00 0.94 0.87 0.91 0.87
1991 1.00 0.84 0.92 0.88
1992 1.00 0.83 0.83
1995 1.00 0.98
1996 1.00

Sources: GUS, RECESS, and authors' calculations.

One could further wonder whether divergent growth performance and prospects have prompted regional migration flows. No such effects could be identified in the demographic data. However, in a context of enduring housing shortages, geographical labor mobility may in some cases involve commuting rather than migrating, at least in the shorter run.

IV. Industrial Output

Given the contribution of industry to overall growth, and the existence of comparatively better data than for the service sector, further disaggregation within industry can provide some relevant insights on the growth process.

Unlike for GDP as a whole, long-run historical series are available for industry, allowing us to put recent developments in a deeper historical perspective. The contraction associated with transition far exceeded any earlier decline, including the large depression of the early 1980s,(18) to the point that the volume of industrial output -- measured in gross terms without adjustment for intermediate inputs -- in 1991 had reverted to its level of the mid-1970s (Figure 5). Growth in 1992-98 was about as vigorous as in the 1960s and the first half of the 1970s, and by 1996, output exceeded its historical 1988 peak. Like for GDP as a whole, growth was distinctly stronger in Poland than in neighboring countries following the transitional collapse (Figure 6).

[GRAPHS OMITTED]

The composition of industrial output shifted considerably over time. At the broadest level, and at constant 1991 prices, the share of mining halved between 1991 and 1998, to 5 percent, whereas that of manufacturing increased by close to 12 percentage points, to some 90 percent. At the same time, the share of the electricity, gas and water supply sector dropped almost as sharply as that of mining. In the case of mining, the drop reflected the fact that the contraction lasted through 1993 and that output fell precipitously in 1998, in a context of weakening coal prices, downscaling of production facilities and rising imports (Table 4). In stark contrast, manufacturing output almost doubled between 1991 and 1998.

Table 4:
Sales Volume in Industry
(Rate of change in real terms)

 1991 1992 1993 1994 1995

Total -8.0 2.8 6.4 12.1 9.7
Mining and quarrying -2.6 -5.3 -4.1 4.6 -0.6
Manufacturing -10.3 5.1 10.4 13.7 11.6
 Food products and beverages 0.6 1.7 8.6 12.8 9.3
 Tobacco products -15.9 6.0 16.6 10.0 -0.5
 Textiles -13.6 3.4 9.9 14.6 -1.0
 Clothing -5.2 15.7 8.3 11.0 2.2
 Leather and leather products -7.7 -5.7 -1.2 12.6 8.2
 Wood and wood products -2.6 15.4 3.7 10.6 10.0
 Pulp and paper -3.9 12.0 6.6 24.7 18.2
 Publishing and printing -15.5 28.4 39.0 6.9 17.3
 Coke, refined petroleum -14.9 12.8 12.3 7.2 5.6
 products and derivatives
 Chemicals and chemical -9.9 -2.0 6.1 17.3 13.1
 products
 Rubber and plastic products 3.0 29.9 19.9 16.1 17.0
 Other non-metallic mineral -3.3 -1.4 9.8 14.6 4.8
 products
 Basic metals -23.8 -2.3 1.8 16.7 15.2
 Metal products (except 1.0 19.8 7.1 15.6 15.8
 machinery and equipment)
 Machinery and equipment -24.7 -7.7 9.4 15.5 20.9
 Office equipment and -35.1 -16.0 51.4 -1.6 52.2
 computers
 Electrical machinery and -6.7 2.1 10.9 9.7 16.1
 apparatus
 Radio, TV and -23.5 14.9 27.0 26.8 17.4
 communication equipment
 Medical, precision and -9.6 3.4 16.1 11.7 25.7
 optical instruments
 Motor vehicles and trailers -22.9 11.0 27.4 13.7 14.1
 Other transport equipment -23.4 7.2 9.6 22.6 3.9
 Furniture and other -6.0 13.0 11.4 14.5 24.2
 manufacturing
 Scrap recycling -15.3 26.8 -9.8 8.0 23.5
Electricity, gas and water 3.5 -4.6 -11.2 4.7 0.9
 supply
 Electricity, gas, steam and 3.3 -5.2 -12.0 4.5 0.8
 hot water supply
 Collection, purification and 6.1 0.5 -2.9 7.0 1.4
 distribution of water

 Share
 Cumul- in
 ative 1998
 1996 1997 1998 1991-98 Sales

Total 8.3 11.5 4.6 70 100.0
Mining and quarrying 2.5 0.5 -13.0 -15 6.1
Manufacturing 9.8 13.4 6.4 95 84.7
 Food products and beverages 9.6 9.6 7.1 75 21.0
 Tobacco products -5.4 4.1 6.2 41 1.0
 Textiles 2.8 6.9 -2.9 38 2.1
 Clothing 4.8 12.7 6.9 79 2.4
 Leather and leather products 12.5 4.2 -15.5 12 0.9
 Wood and wood products 12.7 12.0 9.1 100 2.9
 Pulp and paper 9.8 18.8 8.8 150 1.8
 Publishing and printing 14.4 15.6 14.7 239 2.9
 Coke, refined petroleum 2.8 1.7 -13.0 30 3.3
 products and derivatives
 Chemicals and chemical 4.9 11.9 -2.6 58 5.8
 products
 Rubber and plastic products 17.2 21.1 15.4 247 3.5
 Other non-metallic mineral 9.5 11.9 10.7 76 4.2
 products
 Basic metals 0.0 13.2 -4.8 44 5.5
 Metal products (except 20.4 14.1 19.0 181 4.3
 machinery and equipment)
 Machinery and equipment 9.8 8.7 -1.0 67 5.1
 Office equipment and 44.2 24.3 20.4 311 0.3
 computers
 Electrical machinery and 11.8 15.9 7.7 101 2.7
 apparatus
 Radio, TV and 19.2 29.0 17.6 293 1.8
 communication equipment
 Medical, precision and 14.7 19.7 1.9 136 1.0
 optical instruments
 Motor vehicles and trailers 34.0 31.1 18.3 281 6.0
 Other transport equipment -4.8 11.2 10.2 75 2.2
 Furniture and other 14.2 25.4 18.9 205 3.7
 manufacturing
 Scrap recycling -10.0 17.2 8.1 74 0.3
Electricity, gas and water 0.3 3.4 1.0 -6 9.2
 supply
 Electricity, gas, steam and 0.4 3.5 0.6 -8 8.3
 hot water supply
 Collection, purification and -0.4 3.0 5.9 15 0.9
 distribution of water

Source: GUS.

At a more disaggregated level as well, important shifts were recorded. Looking at the main 23 manufacturing sectors, output grew by at least 5 percent per annum on average in all but three sectors while it boomed in 11 sectors, where annual growth exceeded 10 percent on average (Table 4). While production increased in all 23 manufacturing sectors, it declined or even collapsed for a number of products (Figure 7).(19) The most spectacular declines were recorded for some categories of electronic consumer goods and heavy machinery (radios, metal cutting machines), but strikingly other types of electronics consumer goods and heavy machinery recorded among the most impressive surges (color TVs, heavy vehicles). At the level of individual products, the output mix therefore changed considerably. However, and although some heavy industry items registered declines or very limited growth, the share of heavy industry may not have declined significantly.

[GRAPH OMITTED]

Looking at the behavior over time of the output series for individual products (not shown), it appears that during the period 1992 to 1998 on the whole their degree of synchronization increases with average overall growth, i.e. the dispersion of their growth rates tended to be lower the faster GDP expanded.(20) However, the coefficient of variation of the growth rates of individual items was several times lower prior to 1990, consistent with the idea that transition involves significant modifications of the structure of supply (and demand).(21)

The evolution of output has also varied considerably across regions. Between 1991 and 1998, the volume of industrial output had increased everywhere, more than doubling in 19 of the 49 regions (Figure 8). Growth ranged from 9 to 340 percent, with the five most dynamic regions recording an expansion over 8 times as rapid as the bottom five.

[GRAPH OMITTED]

V. Investment And Productivity

Reverting to a more macroeconomic level of analysis, the question arises as to which forces were underlying the expansion on the supply side. This section documents the shift, over the 1992-98 period, from an intensive to a more extensive growth pattern.

Labor was hoarded on a large scale in Poland in the 1980s. Gora and Rutkowski (1990) for example estimate that labor hoarding amounted to around one fourth of total employment at the time (hoarding being defined as labor that would not have been employed had the economy functioned as a market one with the same level of output). Labor hoarding increased during the 1990-91 contraction, as separations were limited by political and social considerations, so that employment declined less than output. Hence, by the time output started to recover, there was a labor "overhang" on the order of 30 percent. This helps explain why employment continued to decline for another two years, and why it expanded so little from 1994 onwards. In fact, by 1998, real GDP had increased by 42 percent compared with its 1991 trough even as employment was no larger than in 1991.

This would be consistent with full dishoarding by 1998 if the ratio of productively employed labor to GDP was to be the same in 1998 as in 1991. However, there is no compelling reason to believe that this should be the case. Starting from relatively low levels, investment boomed between 1994 and 1998, rising on average by 16 percent per annum in real terms, with a strong foreign contribution (Tables 5 and 6).(22) The increase in investment contributed to quantitatively but also qualitatively rebuild a rather worn out and largely obsolete capital stock. On those grounds, and also given anecdotal evidence of continued overstaffing in some heavy industries, it would seem incorrect to conclude that production factors were fully used by 1998.

Table 5:
Sources of Growth on the Demand Side
(In percent and in real terms)

 Share in 1992 1993 1994
 1991 GDP

GDP 100.0 2.6 3.8 5.2
Consumption 82.0 3.5 4.8 3.9
Private 59.3 2.3 5.2 4.4
Public 22.7 6.4 3.8 2.2
Gross capital formation 19.9 -13.0 12.8 9.1
Fixed investment 19.5 2.3 2.9 9.2
Exports 23.5 10.8 3.2 13.1
Imports -25.4 1.7 13.2 11.3

 1995 1996 1997 1998

GDP 7.0 6.0 6.8 4.8
Consumption 3.2 7.2 6.1 4.1
Private 3.3 8.3 6.9 4.7
Public 2.9 3.4 3.2 1.6
Gross capital formation 10.5 34.0 20.9 13.8
Fixed investment 16.5 19.7 21.7 13.2
Exports 22.8 12.0 12.2 10.3
Imports 24.3 28.0 21.4 13.7

Source: GUS.
Table 6:
Indicators of Foreign Direct Investment
(Flows and stocks)

 1994 1995 1996 1997 1998

Inward Flows (in $ millions)
PAIZ statistics
 Projects of over $1 million 1493 2512 5196 5678 9574
 Estimated total projects ... ... ... 6600 10100
Balance of payments statistics
 Cash-based, gross 542 1132 2768 3077 5129
 Accruals-based, gross 1875 3659 4498 4908 6365
of which:
 purchased shares 884 1807 2845 2663 4323
 reinvested earnings 382 888 244 25 -264
 loans from foreign 397 666 1095 1767 2025
 shareholders
 in-kind contributions 212 298 314 453 281

Foreign direct investment End-period stocks (in $ billions)
 in Poland
PAIZ statistics
 Projects of over $1 million 4.3 6.8 12.0 17.7 27.3
 Estimated total projects ... ... 14.0 20.6 30.7
 Commitments 4.9 5.3 7.9 10.8 13.3
NBP statistics 3.8 7.8 11.5 14.6 22.5
 Of which: equity and 2.8 6.1 8.7 10.1 16.1
 reinvested earnings(1)
Memorandum item: $ GDP2 92.6 126.3 143.0 143.1 157.7

Sources: Polish Agency for Foreign Investment (PAIZ); National Bank of
Poland (NBP).

(1) Includes contributions in kind and excludes loans.

(2) Revised GDP series from 1995 onwards.

Thus, input and output behavior allow us to distinguish two phases in the seven-year expansion. The initial recovery in 1992-93 involved an increase in capacity utilization rates broadly defined, or in other words a move towards the production possibility frontier. Growth took on a more extensive character from around 1994, when factor accumulation picked up, especially as regards capital. The pickup in capital accumulation from 1994 was in part driven by an increase in FDI, which rose from 2 percent of GDP in 1994 to 4 percent of GDP in 1998. As labor productivity is on average higher in firms with foreign involvement,(23) this increase in FDI also boosted labor productivity, in those manufacturing sectors that received large inflows in particular.(24)

Measuring productivity developments in terms of productivity of labor only, at the most aggregate level, labor productivity rose at an annual rate of around 6 percent during the initial recovery, as against 4 percent rate in subsequent years (Table 7).(25) As illustrated in Figure 9, a similar pattern was recorded in Hungary (where the contribution of FDI was even more important), whereas productivity improved much less in the Czech Republic and the Slovak Republic.

[GRAPH OMITTED]

Table 7:
Productivity Developments in the Visegrad Countries
(annual percentage change(1, 2))

 Czech Republic

 Whole Economy Industry

 Y K L TFP Y K L TFP

1980 2.4 4.8 0.7 0.3 1.8 5.3 0.2 -0.1
1981 -0.7 5.2 0.4 -2.9 -0.7 6.3 0.1 -3.0
1982 -0.8 4.3 0.3 -2.5 -1.5 5.0 0.3 -3.4
1983 1.2 4.0 0.1 -0.3 0.8 4.0 0.4 -0.9
1984 2.6 4.3 0.7 0.7 3.9 4.6 0.1 2.2
1985 0.6 4.3 0.8 -1.4 2.2 5.2 0.4 0.2
1986 2.1 4.2 1.1 -0.1 2.6 5.0 1.2 0.0
1987 0.6 3.5 0.4 -0.9 3.2 4.3 0.5 1.4
1988 2.0 3.7 0.5 0.4 3.2 4.3 0.1 1.6
1989 4.4 3.1 0.6 3.0 3.2 3.7 0.4 1.6
1990 -1.2 2.5 -1.0 -1.5 -1.4 2.3 -4.3 0.5
1991 -12.2 2.4 -5.6 -9.4 -23.8 2.5 -3.8 -22.2
1992 -3.4 2.3 -2.6 -2.4 -8.2 3.0 -8.1 -4.1
1993 0.6 2.3 -1.6 0.8 -5.4 2.8 -5.0 -3.2
1994 3.1 1.9 0.8 2.0 2.1 3.3 -5.5 4.5
1995 6.2 3.0 2.6 3.5 5.7 4.0 0.6 3.9
1996 3.8 4.1 0.7 1.9 9.6 3.9 -0.8 8.8
1997 0.3 3.8 -1.0 -0.3 7.7 3.6 -1.0 7.1
1998 -2.4 3.5 1.2 -2.0 -1.6 0.2 1.3 -2.7

 Hungary

 Whole Economy Industry

 Y K L TFP Y K L TFP

1980 0.2 5.0 -0.7 -1.1 -1.1 6.2 -2.4 -1.7
1981 2.9 4.1 -0.7 1.9 4.9 5.1 -2.3 4.6
1982 2.8 4.2 -0.4 1.6 4.6 4.8 -2.4 4.4
1983 0.7 4.2 -0.6 -0.4 1.8 5.5 -2.4 1.4
1984 2.7 4.0 -0.6 1.7 2.5 5.3 -1.2 1.4
1985 -0.3 3.3 -0.5 -1.1 -2.1 3.3 -0.3 -3.1
1986 1.5 3.7 -0.3 0.3 -0.5 5.6 -0.4 -2.2
1987 4.0 3.1 -0.5 3.2 3.2 3.7 -1.3 2.7
1988 -0.1 3.5 -0.6 -0.9 -1.5 3.8 -2.0 -1.5
1989 0.7 6.4 -0.5 -1.2 -2.0 4.3 -2.1 -2.2
1990 -3.6 4.3 -1.6 -4.1 -8.0 3.7 -2.7 -7.6
1991 -12.7 3.6 -6.0 -10.0 -19.6 3.5 -7.7 -15.9
1992 -3.1 3.4 -9.4 1.8 -6.9 3.6 -10.9 -1.1
1993 -0.6 3.4 -6.5 2.4 3.0 3.2 -11.2 9.1
1994 2.9 3.7 -2.0 2.9 5.8 3.5 -4.6 7.6
1995 1.5 3.4 -2.0 1.6 6.7 3.5 -5.5 9.0
1996 1.3 3.4 -0.8 0.6 3.1 3.6 -0.9 2.4
1997 4.5 3.6 0.0 3.3 14.0 3.8 1.7 11.6
1998 5.0 3.9 1.4 2.7 11.2 4.5 4.5 6.6

 Poland(3)

 Whole Economy Industry

 Y K L TFP Y K L TFP

1980 -6.2 4.3 4.6 -10.7 -4.2 4.5 0.2 -5.9
1981 -10.5 3.3 0.5 -12.0 -14.7 3.4 -0.2 -15.8
1982 -4.9 1.9 -2.5 -4.0 -3.9 2.1 -4.9 -1.4
1983 5.4 2.5 -0.3 4.8 5.4 2.8 -0.3 4.6
1984 5.4 2.4 0.3 4.4 5.0 3.1 0.5 3.6
1985 3.5 2.6 0.9 2.1 3.7 3.4 0.1 2.5
1986 4.1 1.9 0.3 3.3 4.2 2.4 -1.9 4.6
1987 2.0 3.8 -0.3 0.8 3.1 3.7 0.2 1.7
1988 4.0 2.5 -0.7 3.6 4.5 2.6 -0.4 3.9
1989 0.2 3.7 -1.0 -0.4 -2.1 3.2 0.0 -3.3
1990 -12.3 1.1 -3.5 -10.5 -24.8 2.3 -5.8 -21.9
1991 -7.9 1.2 -6.0 -4.4 -18.8 0.9 -8.3 -13.6
1992 1.5 1.3 -4.3 3.8 2.6 1.2 -9.1 8.0
1993 3.7 1.9 -2.4 4.6 8.3 1.7 -5.6 11.3
1994 5.1 2.6 1.0 3.5 9.8 2.8 -0.8 9.4
1995 6.8 2.1 1.8 4.9 9.9 0.3 3.1 7.7
1996 5.8 4.7 1.9 2.9 7.3 5.9 -0.7 5.7
1997 6.6 3.4 2.7 3.6 9.8 3.5 0.3 8.4
1998 4.7 4.0 0.3 3.1 4.3 2.1 -3.0 4.8

 Slovak Republic

 Whole Economy Industry

 Y K L TFP Y K L TFP

1980 3.1 6.4 0.7 0.4 4.0 6.9 1.2 0.8
1981 -0.1 6.0 1.1 -2.9 1.5 7.0 1.9 -2.2
1982 -0.2 5.2 0.6 -2.4 -1.7 5.3 0.7 -4.0
1983 3.2 5.2 1.1 0.7 3.8 5.7 1.2 1.0
1984 4.5 5.5 1.4 1.6 6.8 6.3 1.3 3.8
1985 3.8 5.2 1.3 1.2 5.9 6.5 1.3 2.8
1986 4.0 4.5 1.8 1.3 4.4 3.9 1.7 1.9
1987 2.5 4.0 1.1 0.5 4.7 4.1 1.1 2.6
1988 1.9 3.9 1.0 -0.1 3.5 3.9 0.9 1.6
1989 1.1 3.8 -0.2 -0.1 1.9 3.7 -0.8 1.2
1990 -2.5 3.3 -0.8 -3.1 -2.8 3.8 -2.1 -2.8
1991 -15.8 2.7 -8.3 -11.3 -21.6 3.0 -8.3 -17.2
1992 -6.7 2.9 0.2 -7.9 -9.8 3.2 -13.0 -2.4
1993 -3.8 2.9 0.0 -4.7 -3.9 3.1 -5.2 -1.6
1994 4.8 2.8 -1.8 4.9 4.7 3.1 -5.6 7.3
1995 6.7 3.0 2.2 4.2 8.0 3.3 1.0 6.2
1996 6.4 4.0 0.8 4.5 0.4 4.4 -1.5 -0.1
1997 6.3 4.3 0.2 4.6 2.3 4.7 -3.4 2.9
1998 4.3 4.4 1.2 3.0 0.9 0.2 1.5 0.9

Source: National statistical authorities, and authors' calculations.

(1) Based on logarithmic growth rates.

(2) Y stands for output, K for Capital, L for Labor, and TFP for
Total Factor Productivity

(3) Unrevised GUS series.

In manufacturing as well, labor productivity gains have tended to decline somewhat over time, from over 15 percent per year in 1992-93 to a bit less than 10 percent on average in subsequent years. Those are far more rapid increases than during earlier decades, which in industry averaged 8 percent in the 1950s, 5 percent in the 1960s, 6 percent in the 1970s and 4 percent in the 1980s (excluding 1980-81).(26) In cumulative terms, manufacturing output per unit of labor more than doubled between 1991 and 1998.(27)

Similar insights emerge from an analysis that explicitly takes into account the contribution of capital accumulation and computes the rate of growth of total factor productivity (TFP), the component of the rate of output growth that cannot be attributed to increased input of labor or capital accumulation.(28) Aggregate TFP growth turned positive again in 1992, in tandem with the return to growth of the overall economy, and averaged somewhat less than 4 percent on an annual basis during 1992-98. During the initial recovery in 1992-93, the combined contribution of changes in capital and labor remained negative,(29) and positive TFP growth in excess of overall growth mainly reflected an increase in capacity utilization rates. In the following years, as renewed growth in input of factors, in particular capital, was recorded, the contribution of TFP growth to overall growth fell back to somewhat less than two-thirds.

The average rate of TFP growth in Poland over the 1992-98 period has been remarkably high by international standards. Among emerging market economies, comparable TFP growth has only been achieved by China in recent years.(30) And only two industrial countries, Finland and Ireland, registered a growth rate of TFP exceeding 3 percent in the 1990s. The experience of Finland, which entered a deep recession in the early years of the decade as trade with Russia collapsed but then returned to robust output growth in the second half of the 1990s, is akin to Poland's.

Poland's productivity gains since 1992 reflect in part the response to the increasingly competitive environment faced by Polish industry, as trade expanded rapidly in the wake of the 1990 liberalization. From 1992 to 1998, the volume of Polish exports more than doubled, while that of imports nearly tripled. This surge in trade was mainly driven by a rapid expansion in manufacturing trade with the EU countries. The EU share in total Polish exports rose from less than one third in 1992 to almost 70 percent in 1998, while the EU share in total imports rose from around around one half to more than 65 percent in the same period. Manufacturing and machinery products were the fastest growing component of total exports between 1992 and 1998, as their share rose from less than one third to more than one half in this period. The expansion of manufacturing trade with the EU has been a key factor explaining the sharp productivity improvements in Polish industry.

Productivity gains have in part stemmed from the reallocation of inputs of factors across sectors. Employment in agriculture and industry dropped significantly through 1993, even as it was growing in heretofore underdeveloped sectors such as trade (Table 8). Within manufacturing, labor also shifted considerably, as documented by Marchetti (1997). In the process, labor inputs moved out of state-owned enterprises and into de novo firms (Jackson et al., 1999). Inputs of capital were also subject to considerable sectoral reallocation in the period 1993 to 1998, reflecting the uneven distribution of the economy-wide investment boom, with investment recording only modest gains in agriculture but expanding rapidly in trade and services activities. Decomposing the economy into six broad sectors (agriculture, industry, construction, transport and communication, trade, and other services), the sectoral reallocation of inputs accounted for around one-fourth of the aggregate gains in TFP during the 1992-98 period of renewed growth. This is almost twice as much as its contribution during the 1983-89 period, the last episode of pre-transitional growth.(31)

Table 8:
Employment
(In percent)

 Shares
 1989(1) 1992(1) 1992(2) 1993(2)

Agriculture and 26.7 26.2 25.8 25.7
 forestry
Industry 28.8 26.5 25.8 25.6
Mining 2.9 3.1 3.1 2.9
Manufacturing 25.3 22.9 21.0 20.8
Construction 7.8 7.3 7.0 6.0
Services 36.7 40.1 41.6 42.7
Trade 8.6 10.9 12.3 13.1
Transport and 5.8 5.3 6.2 6.1
 communication
Finances 1.0 1.4 1.3 1.5
Public administration 1.5 2.0 2.2 2.4
Health and education 11.5 13.0 12.2 12.5
Total 100.0 100.0 100.0 100.0

 Change
 1998(2) 1989- 1993-
 93(1) 98(2)

Agriculture and 25.7 -17.4 8.0
 forestry
Industry 23.4 -25.8 -1.1
Mining 2.0 -16.4 -27.3
Manufacturing 20.7 -26.9 7.5
Construction 5.8 -34.7 4.6
Services 45.1 -2.5 14.2
Trade 13.6 33.5 12.3
Transport and 5.6 -24.6 -0.5
 communication
Finances 2.0 29.8 41.3
Public administration 2.7 26.8 25.3
Health and education 12.2 -3.5 5.6
Total 100.0 -15.7 8.1

Sources: GUS, and authors' computations.

(1) Former classification.

(2) SNA classification.

TFP computations at the sectoral level indicate that in industry TFP growth also exceeded output growth during the first two recovery years, as industrial employment continued to shrink. Thereafter and through 1998, the contribution to output growth from increases in TFP relative to factor accumulation was more important in industry than in the overall economy, reflecting an only modest pick-up in employment in the sector.(32)

VI. Sustainability

The sustainability of Poland's high-growth performance started to be questioned in 1998, as the economy showed signs of deceleration. The slowdown was partly caused by a deliberate tightening of financial policies intended to restrain domestic demand and to contain a widening current account deficit. It also reflected external factors, namely the slowdown in Western Europe and the shock imparted by the sharp decline in demand from Russia and other countries east of Poland, illustrating the sensitivity of growth to exogenous developments in export markets. As a result, real GDP growth was down to 2.9 percent in the last quarter of 1998 compared with a year earlier, slowing further to only 1.6 percent in the first quarter of 1999, with gross industrial output shrinking in those two quarters for the first time since 1991. But the growth slowdown proved to be temporary and GDP growth picked up again by the fourth quarter. This episode has raised questions, however, as to whether strong growth is sustainable and what Poland's medium-term growth prospects are.

Poland's growth potential can be assessed using the coefficients from growth regressions derived from the neoclassical growth model and estimated for a large sample of market economies. Regressions that try to properly address measurement problems related to initial per capita income gaps and human capital and that add an institutional quality variable indicate that Poland can achieve average growth rates of per capita income at 5.7 percent per year over a longer-time horizon (Crafts and Kaiser, 2000).(33) These results also suggest that the robust 1994-98 growth performance was broadly in line with potential. Even with sustained high growth, convergence towards EU per capita income levels is bound to be a drawn out process. Based on the per capita income gap in 1998 and assuming that per capita income growth will on average be 5.7 percent in Poland and 2 percent in the current EU members, a full catch-up will require almost 30 years.

Longer-term growth projections can also be derived from a growth accounting exercise. In this approach, Poland's growth potential is based on a combination of an increase in the inputs of factors and TFP gains. Moderate growth in the size of Poland's working age population together with further improvements in the quality of human capital are projected to result in an increase in inputs of labor, measured in effective terms, by 1 percent per year. Assuming that gross fixed capital formation will be maintained at its 1998 level of about 25 percent of GDP and that 8 percent of the capital stock will be scrapped every year, inputs of capital are projected to grow at an annual rate of around 5 3/4 percent.(34) Under these assumptions regarding factor inputs, high and sustained TFP growth of 3 percent per year will still be needed for the 5.7 per capita growth projection to materialize. A more modest increase in capital inputs or lower TFP growth would be associated with significantly weaker longer-term growth.(35)

Poland's growth potential is strong, provided the high investment rates and robust TFP growth recorded in recent years can be sustained. In view of relatively low domestic savings rates, covering the country's investment needs will likely require continued access to foreign savings, FDI in particular. The scope for continued robust TFP growth to a large extent depends on the choice of economic policies. Supply-side policies need to address persisting infrastructure bottlenecks, which cause congestion and may inhibit growth.(36) This is particularly conspicuous in the case of the road network, which is quantitatively and qualitatively lacking (European Commission, 1998). Further institutional reform is also needed, and it is likely to be stimulated by the efforts to gain EU accession. Weaknesses in the judiciary system -- where the average time for processing cases has tended to lengthen and enforcement of court rulings often remains wanting -- slow down or altogether discourage certain business ventures but also facilitate corruption, which has been shown to be growth-inhibiting (Mauro, 1996). Also, ambitious institutional reforms launched in 1999 in such areas as devolution, pensions, and health care have to be implemented fully. Additional productivity gains will result from continued reallocation of capital and labor from low- to high-productivity sectors. Further downsizing in agriculture and heavy industry, in particular, is expected to reduce these sectors' sizable drag on overall growth.

VII. Main Conclusions

This paper has highlighted some of the features of Poland's growth performance in the 1990s. This performance has been quite commendable compared with neighboring countries, but it has not been exceptional by the standards of rapid growth episodes in other regions. First, growth has been largely pulled by manufacturing, reflecting a recovery from the sharp 1990-91 contraction in industry initially and the contribution from a pickup in factor accumulation later on. Second, transition has involved wholesale shifts in the composition of industrial output. These shifts, which have contributed to considerable productivity gains in industry, have been stimulated by a trade reorientation toward Western Europe and large FDI inflows. Third, even though growth has been broad-based, it has been uneven across regions. In a context of limited geographical labor mobility, this may have implications as regards the speed with which structural reforms can be implemented.

In terms of policies, Poland's performance illustrates the crucial importance of letting the reallocation of capital and labor run its course. This requires sound macroeconomic policies, so as to instill confidence with potential investors, coupled with liberalization on the structural front. Such policies create an environment conducive to the necessary redistribution of existing resources and to the attraction of external ones in the form of FDI.

Notes

(1.) Mark De Broeck is affiliated with the IMF, and Vincent Koen with the OECD. The views in this paper are the authors' alone and should not be taken as representative of the IMF or the OECD. The authors are grateful to Witold Orlowski for sharing regional GDP data, and to Tam Bayoumi, Marek Belka, Andrew Berg, Witold Orlowski, and Nancy Wagner for useful comments. All interpretations and remaining errors are the authors' own.

(2.) In 1998, the Central Statistical Office (GUS) started to publish a new GDP series, running from 1995 onwards, but with only minute changes for the real GDP index (GUS, 1998). Therefore, the aforementioned average and cumulative figures hold both for the old and the new series.

(3.) See, for instance, Kolodko (1989) and Sachs (1993).

(4.) No consistent long-run GDP series are available, but indices of sectoral developments, and in particular industrial output (see below), support this claim.

(5.) The series shown in Figure 2 reflects revisions by GUS around the mid-1990s, including an 8 percent (rather than 11.6 percent) real GDP decline in 1990. To maintain consistency with the employment and capital stock data, the unrevised growth series (in log terms) is used to derive the productivity calculations in Table 7.

(6.) Recently, an alternative series was published showing lower growth in the late 1980s and drops on the order of only 5 to 6 percent both in 1990 and in 1991, implying that GDP may have recovered to its 1989 level already in 1994 (RECESS, 1999). This alternative series vindicates early analyses suggesting that the official data vastly exaggerated the magnitude of the contraction (Berg and Sachs, 1992).

(7.) For analyses of the factors underlying Poland's strong performance, see among others Balcerowicz and Gelb (1994), Balcerowicz (1995), Borensztein and Ostry (1992), Gomulka (1998), Johnson et al. (1999), Kolodko and Nuti (1997), Lane (1992), OECD (1996), Pinto et al. (1993), and World Bank (1994).

(8.) Or even 27 percent, based on the revised national accounts series, which lifts the level of nominal GDP by around 6 percent.

(9.) Or even 57 percent, based on the revised national accounts series.

(10.) Market services include the trade and repair; hotels and restaurants; transport, storage and communication; financial intermediation; real estate and business activities; and other categories, see Table 1.

(11.) The state sector continues, however, to account for the bulk of both employment and output in the mining sector.

(12.) See Dabrowski, Gomulka, and Rostowski, 2000.

(13.) Unless specified, the term "regions" henceforth denotes the 49 "voivods" under the system prevailing until 1998 (administrative reform in 1998 reduced the number of regions to 16 and created a new layer of counties ("powiats") between regions and municipalities ("gminas").

(14.) See RECESS (1994). The 1986, 1990 and 1991 GDP estimates were kindly provided by Dr. Orlowki of RECESS. They are not based on exactly the same methodology as the subsequent ones, which mainly affects the ranking of the Warszawkie region, the capital's weight being higher, all else equal, under the former than under the new methodology. Moreover, in the absence of regional GDP deflators, inferences about regional developments are drawn on the admittedly problematic assumption that over the pluri-annual spells considered inflation has been relatively uniform across areas.

(15.) In order to ensure data comparability, the base is 1992 rather than the 1991 national trough.

(16.) Since the capital is most affected by the change in methodology between 1991 and 1992, Table 2 shows results including and excluding it from the sample.

(17.) The report of the Task Force for Regional Development in Poland (1996) discusses which regions gained and which ones lost most through 1994.

(18.) On that episode, see for instance Hanson (1982).

(19.) The products shown in Figure 7 are the main items for which GUS monitors output and for which information is available over a long time span. In that sense, they constitute a representative set.

(20.) This is a striking empirical regularity save for 1995.

(21.) In this computation, growth rates are expressed as indices with the previous year set equal to 100, in order to preserve the relevance of the coefficient of variation as a measure of dispersion when the mean approaches zero.

(22.) For a detailed analysis of foreign direct investment flows, see FTRI (1998). In 1995-98, foreign direct investment measured on an accruals basis accounted for about 15 percent of gross fixed capital formation.

(23.) Konings (1999) presents panel data evidence on this productivity differential.

(24.) Data on cumulative FDI flows through 1998, as published by the Polish Foreign Investment Agency (PAIZ), show an uneven distribution across the 10 manufacturing sectors for which information is available. Labor productivity gains have been particularly strong in manufacturing sectors with a high contribution of FDI to overall capital formation, as reflected in a correlation coefficient of 0.71 for the period 1992 through 1998 for the 10 sectors.

(25.) As illustrated in Figure 9, a similar pattern was recorded in Hungary (where the contribution of FDI was even more important), whereas productivity improved much less in the Czech Republic and the Slovak Republic.

(26.) Including that recession in the sample brings down the average productivity increase to 2 percent for the 1980s.

(27.) An equally impressive surge was recorded in Hungary over the same period. Again, the Czech Republic and the Slovak Republic witnessed more limited productivity gains.

(28.) In this context, TFP growth should be regarded as truly a residual, reflecting the effect of a variety of factors that influence the efficiency with which factors of production are used, and should not be interpreted as an exogenous rate of technological progress.

(29.) In these TFP calculations, the weight associated with capital is 0.35 and that with labor 0.65.

(30.) Based on the numbers presented in Collins and Bosworth (1996).

(31.) The calculations are based on the approach followed by Bernard and Jones (1996) and Cameron et al. (1997). These authors compute the contribution of sectoral reallocation of inputs to overall productivity growth by decomposing the change in aggregate TFP into a productivity change effect and a reallocation effect. The first effect measures the contribution of productivity changes within each sector, and the second one the contribution of changes in sectoral composition.

(32.) A comparison with the Czech Republic, Hungary, and the Slovak Republic reveals broadly the same picture. TFP growth initially exceeded output growth, indicating an increase in the rate of capacity utilization, and subsequently made the largest contribution to output growth, especially in industry. The computations are based on the capital stock data published in the national statistical yearbooks, extrapolated using investment data for the most recent years. For Hungary, Darvas and Simon (1999) have derived an alternative set of computations based upon a vintage model to reconstruct the capital stock data.

(33.) These authors also discuss the robustness of earlier estimates in Fischer et al., 1998. For growth regressions that only cover the transition economies as such, see Berg et al., 1999.

(34.) Projected capital stock growth also reflects the assumed initial capital-output ratio. Based on survey data from the Central Statistical Office, the ratio is taken to be 1.7 in 1998.

(35.) The growth projection is also sensitive to the assumptions regarding the initial capital-output ratio and the depreciation rate. Part of the central planning-era capital stock is likely to be obsolete, and this effect can be captured by either reducing the initial capital-output ratio or -- as in the growth scenario here -- assuming a relatively high depreciation rate.

(36.) See Canning (1998) and the references therein for a cross-country approach.

References

Anderson, Ronald, and Kegels, Chantal, Transition Banking: Financial Development of Central and Eastern Europe. Oxford: Clarendon Press, 1998.

Balcerowicz, Leszek, Socialism, Capitalism, Transformation. Budapest: Central European University Press, 1995.

Balcerowicz, Leszek, and Gelb, Alan, "Macropolicies in Transition to a Market Economy: A Three-Year Perspective." Proceedings of the World Bank Annual Conference on Development Economics, 1994.

Berg, Andrew, "Does Macroeconomic Reform Cause Structural Adjustment? Lessons From Poland." Journal of Comparative Economics 18, 3: 376-409, June 1994.

Berg, Andrew, Borensztein Eduardo, Sahay, Ratna, and Zettelmeyer Jeromin, "The Evolution of Output in Transition Economies: Explaining the Differences." IMF Working Paper No. 99/73, May 1999.

Berg, Andrew, and Sachs, Jeffrey, "Structural Adjustment and International Trade in Eastern Europe: The Case of Poland," Economic Policy 14:117-73, April 1992.

Bernard, Andrew, and Jones, Charles, "Productivity Across Industries and Countries: Time Series Theory and Evidence." Review of Economics and Statistics 78, 1: 135-146, February 1996.

Borensztein, Eduardo, and Ostry, Jonathan, "Structural and Macroeconomic Determinants of the Output Decline in Poland: 1990-91." IMF Working Paper No. 92/ 86, October 1992.

Bratowksi, Andrzej, Grosfeld, Irena, and Rostowski, Jacek, "Investment and Finance in De Novo Private Firms. Empirical Results from the Czech Republic, Hungary, and Poland." Economics of Transition 8, 1:101-116.

Cameron Gavin, Proudman, James, and Redding, Stephen, "Deconstructing Growth in UK Manufacturing," Bank of England Working Paper No. 73, December 1997.

Campos, Nauro, "Context is Everything. Measuring Institutional Change in Transition Economies," World Bank Policy Research Working Paper No. 2269, January 2000.

Canning, David, "A Database of World Stocks of Infrastructure, 1950-95." The World Bank Economic Review 12, 3: 529-47, September 1998.

Collins, Susan, and Bosworth, Barry, "Economic Growth in East Asia: Accumulation versus Assimilation." Brookings Papers on Economic Activity 27, 2:135-191, 1996.

Crafts, Nicholas, and Kaiser, Kai, "Long Term Growth Prospects in Transition Economies: A Reappraisal." Mimeo, London School of Economics, September 2000.

Czyzewski, Adam, Orlowski, Witold, and Zienkowski, Leszek, "Poland: Optimistic Scenario of Economic Development until 2005." RECESS Research Bulletin 3, 2:17-41, 1994.

Dabrowski, Marek, Gomulka, Stanislaw, and Rostowski, Jacek, "Whence Reform? A Critique of the Stiglitz Perspective." Centre for Social and Economic Research Working Paper, March 2000.

Darvas, Zsolt, and Simon, Andras, "Capital Stock and Economic Developments in Hungary," National Bank of Hungary, Working Paper No. 1999/3, May 1999.

De Broeck, Mark, and Koen, Vincent, "The Great Contractions in Russia, the Baltics and the Other Countries of the Former Soviet Union: A View from the Supply Side," IMF Working Paper No. 00/32, March 2000.

Deichmann, Uwe, and Henderson, Vernon, "Urban and Regional Dynamics in Poland," World Bank Policy Research Working Paper No. 2457, September 2000.

European Commission, Regular Report from the Commission on Poland's Progress Towards Accession. Brussels: European Commission, 1998.

Fischer, Stanley, Sahay, Ratna, and Vegh, Carlos, "How Far is Eastern Europe from Brussels?," IMF Working Paper No. 98/53, April 1998.

FTRI, Foreign Investment in Poland, Warsaw: Foreign Trade Research Institute, 1998.

Gomulka, Stanislaw, "The Polish Model of Transformation and Growth," Economics of Transition 6, 1: 163-71, May 1998.

Gora, Marek and Rutkowski, Michak, "The Demand for Labour and the Disguised Unemployment in Poland in the 1980s," Communist Economies, 2, 3: pp. 325-34, 1990.

GUS, 1998 Statistical Yearbook, Warsaw: Central Statistical Office, 1998.

Hanson, Philip, "The End of Import-Led Growth? Some Observations on Soviet, Polish and Hungarian Experience in the 1970s" Journal of Comparative Economics, 6, 2: 130-47, June 1982.

International Monetary Fund, World Economic Outlook. Washington: IMF, October 1996.

Jackson, John, Klich Jacek, Poznaska, Krystyna, and Chmiel, Jozef, "The Continued Importance of Business Creation: The Dynamics of the Polish Economy 1990-1996," RECESS Research Bulletin 8, 1:5-40, 1999.

Johnson, Simon, McMillan, John, and Woodruff, Christopher, "Contract Enforcement in Transition," CEPR Discussion Paper No. 2081, February 1999.

Kennedy, Robert, "A Tale of Two Economies: Economic Restructuring in Post-Socialist Poland," World Development, 25, 6: 841-65, June 1997.

Kolodko, Gregorz "Stabilisation Policy in Poland. Challenges and Constraints," Institute of Finance Working Paper No. 3, 1999.

Kolodko, Gregorz, and Nuti, Domenica Mario, "The Polish Alternative. Old Myths, Hard Facts and New Strategies in the Successful Transformation of the Polish Economy," WIDER Research for Action No. 33, 1997.

Konings, Jozef, "The Effect of Direct Foreign Investment on Domestic Firms: Evidence from Firm Level Panel Data in Emerging Economies," LICOS Discussion Paper 86/ 1999, October 1999.

Landesmann, Michael, "Structural Change in the Transition Economies," Vienna Institute for International Economic Studies Working Paper No. 269, September 2000.

Lane, Timothy, "Inflation Stabilization and Economic Transformation in Poland: The First Year." Carnegie-Rochester Conference Series on Public Policy 36:105-56, July 1992.

Marchetti, Domenico, Productivity and the Business Cycle, Three Essays on U.S. and Polish Manufacturing. New York: Garland, 1997.

Mauro, Paulo, "The Effects of Corruption on Growth, Investment and Government Expenditure," IMF Working Paper No. 96/98, September 1996.

Ministry of Economy, Poland's Report: Industry in 1997. Warsaw: Ministry of Economy, 1998.

Organisation for Economic Co-operation and Development, OECD Economic Surveys 1996-1997: Poland. Paris: OECD, 1996.

Organisation for Economic Co-operation and Development, National Accounts: Main Aggregates 1960-1997. Paris: OECD, 1998.

Pinto, Brian, Belka, Marek, and Krajewski, Stefan, "Transforming State Enterprises in Poland: Evidence on Adjustment by Manufacturing Firms," Brookings Papers on Economic Activity 24, 1: 213-70, 1993.

Research Centre for Economic and Statistical Studies of the Central Statistical Office and the Polish Academy of Sciences, "New GDP Estimates 1985-1995. (In Polish)" Studia I Prace, No. 263, 1999.

Research Centre for Economic and Statistical Studies of the Central Statistical Office and the Polish Academy of Sciences, and Statistical Office in Katowice, Gross Domestic Product by Voivodships for 1995: Experimental Calculations. Warsaw: RECESS, August 1997.

Research Centre for Economic and Statistical Studies of the Central Statistical Office and the Polish Academy of Sciences, "Spatial Differentiation of Gross Domestic Product per capita in Poland: 1986-1992." RECESS Research Bulletin 3, 2:3-15, 1994.

Sachs, Jeffrey, Poland's Jump to the Market Economy. Cambridge, Massachusetts: MIT Press, 1993.

Strategy for Poland, mimeo, Warsaw, June 1994.

Task Force for Regional Development in Poland, Regional Development in Poland 1990-95, Diagnostic Report. Warsaw, 1996.

World Bank, Poland: Policies for Growth with Equity. Washington DC: World Bank, 1994.

JEL Classification Numbers: O47, P24

Mark De Broeck
IMF

Vincent Koen(1)
OECD