Part One of this article published in the March edition of the Bulletin reviewed the claims by the World Bank, European Bank and IMF that there was a strong correlation between the implementation of reform policy and economic growth, and that on the whole transition was working well, where consistently applied. Such claims were based on grouping the transition countries according to their degree of economic liberalisation or reform progress, and ignoring two fundamental facts: 1) that the economies facing the simplest transition challenge were the more wealthy economies of former Eastern Europe (FEE); and 2) that the economies of the former Soviet Union (FSU) facing the most difficult transition challenge were politically unable to apply further measures that were demonstrably having such disastrous consequences for their economies. Part One of this article demonstrated that the apparently high correlation between application of reform policy and economic growth for the FSU and FEE economies in the data used by the World Bank, European Bank and IMF did not apply to the FSU economies when considered separately, and was partly a consequence of comparing the FSU with FEE.
In my previous article, a year ago, the economies in transition were grouped together and analysed according to their level of economic decline. In this article the same principle has been retained, but the economies are also ranked according to their level of average wealth (in 1996). These two rankings are very similar partly because the declines have been so great, but mainly because the poorest countries of the FSU and FEE have been the ones which have found it most difficult to undergo transformation, and which have suffered the most.
Appendix 1 (see hard copy of Bulletin only) contains a summary table which has been updated to include the final assessments of 1995 growth, and the preliminary results for 1996.
In this table the Transition Economies are arranged according to rank order of GDP decline in chronological time. The different series of data as presented by EBRD, IBRD and the CIS Goskomstat have been compared. The most reliable series appears to be the latest IBRD series. As noted in the text above the revisions that the Russian Federation Statkom made to its evaluation of the Russian dynamic were made in September 1995 and have been duly incorporated into the current CIS Goskomstat series and the latest IBRD and EBRD series.
For the sake of comparability and comprehensibility, appropriate data for the Transition Economies of Asia (TEA) China, Vietnam and Mongolia have been included in the Appendix, but they are not discussed separately in the text. The FSU and FEE are considered separately and in this order. Within the FSU four separate groups of countries are identified: i) the five Warring States of Georgia, Tajikistan, Azerbaijan, Armenia and Moldova; ii) the four other Central Asian Republics: Kyrgyz Republic, Turkmenistan, Uzbekistan and Kazakhstan; iii) the three Baltic Republics of Lithuania, Latvia and Estonia, and iv) the Slavic Heartland: Ukraine, Belarus and Russia. Within the FEE, three separate groups of countries are identified: i) the four Warring States(1) of Bosnia-Hercegovina, Yugoslavia (Serbia-Montenegro), FYR Macedonia and Croatia, ii) three less developed countries of former Eastern Europe (LDPFEE): Bulgaria, Albania and Romania; and finally iii) the five more developed countries (MDPFEE): Slovak Republic, Hungary, Czech Republic, Poland and Slovenia.
Within each sub-group the countries are ranked according to the level of their 1996 per capita GNP in US$ in Purchase Power Parity.(2) There is a very distinct relationship between per capita income levels between groups and subgroups. The 290 million people in the FSU have since 1989 experienced on average a 33% decline in real GNP from a level roughly equivalent to 3,594US$ per year in 1989 to 1,837US$ per year in 1996, when calculated using the World Bank Atlas exchange rates or from 6,194US$ in 1989 (about 25% of the average US Income) to 3,289US$ in 1996 (about 14% of the average US Income) when calculated in terms of Purchasing Power Parity.(3) As noted above this level was still falling in 1996, although it is hoped that 1997 will register the first overall reversal in these growth rates. During the same period the 123 million people in the FEE suffered a smaller fall in real GNP, which has already been partly recovered so that by 1996 real GNP was only 7.8% below the 1989 level. In terms of US$ equivalence the Atlas exchange rate method would show a fall from 2,738US$ in 1989 to 2,591US$ in 1996. These calculations, which indicate that the average FSU income was more than double the average FEE income in 1989 and that the differential would have fallen to the FSU income being 50% higher in 1996, would not be acceptable to many observers who have observed FEE and FSU living standards. The PPP index is much more reliable here, and it would indicate a fall in FSU GNP from 6,194US$ per head in 1989 to 3,289US$ per head in 1996, while FEE (4) GNP fell from 6,404US$ to 6,031US$ per head, i.e. from a level in which FEE was 5% higher than FSU in 1989 to one in which it was 80% higher in 1996.
During the same period Chinese GNP rose by 89% from 328US$ per capita in 1989 to 619US$ per capita in 1996 measured by Atlas, and from 1,553US$ per capita in 1989 to 2,932US$ per capita measured in more comparable PPP. The FSU can consequently be seen to be rapidly converging with Chinese PPP while FEE remains more than double the Chinese level.
Below we will consider these regions and countries in more detail.
1) Within the FSU
All the countries of the FSU have undergone drastic structural readjustment as a result of: i) the Russian decision to remove previous subsidies on inter-republican trade; ii) the collapse of the Soviet Union; and iii) the freeing of most Russian domestic prices. The 26-fold increase in domestic prices that was unleashed in Russia in 1992 is now largely under control, both in Russia and in most other countries of the FSU. But GNP has fallen to 53% of the 1989 level and was still falling by 6% in 1996. Capital investment has fallen to 25% of its previous pre-transformation level. Stabilisation is being achieved, but there are few signs that it is being accompanied by new investment and that growth will follow. It seems most likely that stabilisation will be accompanied by stagnation for a few more years before production slowly begins to pick up. And within this generally depressing picture there are and will continue to be great regional differences. The worse cases are those of the five warring states in which economic decline has been accompanied by massive military and political disruption.
a) The Five Warring States, where GDP is reportedly 20-40% of 1989 levels
The five Warring States of the FSU with a population of 26.8 million, represent some of the poorest regions of the FSU. Overall they appear to have experienced a 70% decline in GNP since 1989 and continued to experience an overall 1.2% decline in 1996. Their level of economic decline in 1996 was on the whole much lower than for the other regions of FSU. But here again there were great differences in the situations in the different countries.
i) Georgia (5.4 million population). Because of the enormous scale of economic decline in Georgia (with current GNP reported a mere 20% of the 1989 level), Georgia must have the lowest level of real GNP of all the FSU crisis zones. However, because the difficult political situation no reliable figures are available in terms of GNP in comparison with other countries either in the Atlas method or PPP. In recent years there do appear to have been some signs of the beginning of an improvement (a growth by 8-12% in GDP in 1996). It is highly likely that the scale of under-reporting in this economy is large and that there has been a distinct shift to unregistered household economy activities (e.g. growing potatoes on household plots).
ii) Tajikistan (5.8 million population) with a per capita GNP in 1989 of 795$US (Atlas method) and 2,142US$ in PPP, was at that time by far the poorest Republic of the USSR. It has remained the poorest country in the FSU. In 1996 when Tajikistan GNP had fallen to 37% of its 1989 level its per capita GNP (Atlas method) would be about 293$US and about 789$US in PPP, a level significantly lower than Bangladesh, Haiti and Uganda.(5) There are few signs of any improvement in Tajikistan, and because of the continuing sharp deterioration in the economic situation in Tajikistan, with a further decline of 18% in GNP recorded in 1996, Tajikistan has replaced Azerbaijan as the Transition Economy with the second largest decline, 28-35% of 1989 GNP. (The country with the largest decline remains Georgia - see above). For some reason that does not appear clear the IBRD registers a much greater fall for Tajikistan than do the other agencies.
iii) Azerbaijan (7.5 million population) appears to have registered a slight growth in 1996, despite earlier appearances. It remains however the FSU country with the second lowest level of GNP after Tajikistan estimated for 1996 at 442$US per capita (Atlas method) and 1,216$US in PPP. In relative terms the decline in Azerbaijan's level of GNP in 1996 appears to have been 27-36% of its 1989 level and it thus moves from second to the third lowest, slightly ahead of Tajikistan. Again the IBRD provides significantly lower evaluations.
iv) Armenia (3.7 million population) continues to steadily grow from the catastrophic decline that it experienced in 1990-1993. It now registers about 40% of its 1989 GDP level, with a 1996 level per capita GNP of 656$US (Atlas method) and 2,085$US in PPP.
v) Moldova (4.4 million population). The latest EBRD evaluations continue to project a growth of 4% in GDP for Moldova in 1996, but Statkom indicates a decline of 4.6%. It consequently registers between 36 and 42% of its 1989 level. If the EBRD are correct the Moldova per capita GNP (Atlas method) would be about 990$US, if Statkom are correct it would be about 900$US.
The rest of the FSU registered a 45% decline in GNP by 1996 in comparison with 1989, but the results for 1996 alone registered a decline of 4.9% which was significantly worse than for the five Warring States.
b) The Central Asian Republics (excluding Tajikistan)
The Central Asian Republics including Tajikistan were traditionally the poorest regions of the USSR. However with the exception of Tajikistan they appear to have suffered a significantly lower level of economic decline (on average 35%) than all other regions of the FSU. They have consequently improved their position in comparison with most of these other countries, and are now richer in terms of per capita GNP in PPP than the crisis zones of the Transcaucasus discussed above.
i) The Kyrgyz Republic (4.5 million population), which was the second poorest Republic in the USSR in 1989 appears to be the best FSU performer in 1996 and now registers 50-55% of 1989 level GDP. The 1996 per capita GNP for the Kyrgyz Republic would be about 651$US (Atlas method) and 1,788$US PPP.
ii)Turkmenistan (4.4 million population) presents the most problematic picture, and as explained in the previous article, Turkmen statistics appear to be the least reliable. They appear to be pricing oil production in something other than fixed 1989 prices. The CIS Statkom and EBRD have now stopped publishing what are undoubtedly very misleading figures. We need not consequently be too concerned that Turkmenistan was claiming one of the lowest falls in GDP with the lowest Economic and Reform Progress indicators.
iii) Uzbekistan (22.4 million population). The relatively good performance indicators in Uzbekistan, albeit still with poor reform progress indicators, are probably correct since oil and gas production have remained high. While GDP continues to decline slowly, it is now reported to be only 82-84% of 1989 levels. In the scenario of the FSU crisis economies this appears as a striking success story! Per capita GNP (Atlas method) is about 939$US and in PPP it is probably about 2,318$US.
c) The Baltic Countries
In 1989 the Baltic States were the richest regions of the USSR, however they have experienced a much higher decline in GNP than in all areas of the FSU except the five warring zones with 1996 GNP about 49.4% of 1989 levels. Consequently they now appear poorer in terms of per capita GNP than the Slavic heartland countries (see below).
i) Lithuania (3.7 million population). The CIS Statkom indicators for Lithuania are almost certainly highly incomplete. But despite three years of slight growth the IBRD & EBRD indicators of a level of GDP at 40-46% of the 1989 level are probably comparable with data from other Transition Economies. It should be noted that Lithuania has relatively high Economic Liberalization and Reform Progress indicators, that are of the same order as for the other Baltic States of Estonia and Lithuania, but its economic recovery has been far worse. For simple geographical reasons the Lithuanian economy was always much more closely integrated into the former USSR economy, than were the economies of the other Baltic States and the recovery path is again tending to show that in these early stages (at least) growth is more closely dependent upon the initial circumstances and the different levels of structural challenge to transformation, than to the simple application of IBRD or EBRD-approved reform programmes. Lithuanian per capita GDP in 1996 was probably about 1,413$US (Atlas method) and 3,443$US in PPP.
ii) Latvia (2.5 million population). The Latvian economy appears to be stagnating at a level of 50-52% of its 1989 level GDP. Despite lower levels of Reform Progress and Economic Liberalization indices its economy is performing better than the Lithuanian one. Latvian per capita GNP in 1996 was probably 2,306$US (Atlas method) and 3,200$US in PPP.
iii) Estonia (1.5 million population). Estonia is reported to have experienced a positive, but declining growth of GDP for the last three years, following the enormous decline in 1991-93. It now presents the best level of recovery in the European FSU and the Baltic countries at 60-71% of 1989 levels of GDP, which is approaching the lowest non-Warring FEE level (i.e. Bulgaria). Although Estonia's reform progress indicators are the highest in the FSU, they are not really much higher than those for Lithuania, which has experienced a very different and less satisfactory transition path so far. Estonian per capita GNP in 1996 was probably about 2,997$US (Atlas method) and 4,794$US in PPP, which would make Estonia the richest country in the FSU.
d) The Slavic Heartland
The Slavic heartland of Ukraine, Belarus and Russia with a combined population of 211 million dominates the overall picture for the FSU.
i) Ukraine (51.9 million population). The position in Ukraine following a further 10% fall in GDP in 1996 now exceeds the level of GDP decline registered in the best of the Warring States (about 40% of 1989 levels). It might well be that some production is not adequately registered and that a correction similar to that which has been applied to the Russian Federation is needed (see below). In terms of comparability with the Russian data, it most certainly is needed. But a 5-10% adjustment in these levels will not materially affect the situation, which is becoming increasingly grave. Ukrainian per capita GDP in 1996 was about 1,567$US (Atlas method) and 2,149$US in PPP.
ii) Belarus (10.4 million) registered a slight growth of 2% in 1996 which brings it up to 55-60% of its 1989 level GDP. Belarus per capita GNP in 1996 was about 1,843$US (Atlas method) and 3,685$US in PPP. Excluding Estonia, Belarus remains the second richest country in the FSU after Russia.
iii) Russia (148.4 million) With the exception of Estonia, Russia is the richest country in the FSU with a per capita GNP in 1996 of 2,468$US (Atlas method) and 4,293$US in PPP. However, unlike many of the other countries it continued to experience a significant further decline of 6% GDP in 1996. This was a level of decline which was larger than that experienced in 1995. This is despite the fact that its 1996 Economic Liberalization Indicator was very high, 6.9, in comparison with the 1989-95 level of 3.8. On the currently accepted indicators Russia has a level of GDP that is 58% of the 1989 level. For a huge economy with 150 million people to continue to experience major economic declines for seven years, and to reach levels well below those temporarily reached in WW2 or the Great Depression indicates that we are witnessing the most serious crisis the modern world has ever seen. However, the community of analysts seems to be far more concerned with the realisation that a decline to 58% of the 1989 level looks far less worrying than the decline to 49% of the 1989 level that was registered in 1995 before major upward corrections were made to the data. As mentioned above, I am not totally convinced that such revisions were necessary. I do not want to fetishise the nature of statistics; they are always an imperfect reflection of reality, and we have to make do with imperfect reflections, to some extent. It is however disturbing to see major analysts acting as though they expected statistics to be perfect, and then dismissing inconvenient evidence that is reflected in such indicators, on the grounds that they are unreliable. This subject needs a far more serious treatment than it has so far received. Fortunately, the preliminary indicators for 1997 show that the end of the fall has probably arrived. However, it remains unclear whether the end of the fall will mark the beginning of a recovery. Since the level of investments in 1996 was reported to be only 26% of the 1988 level, a rapid increase in production would appear highly unlikely.
2) Within FEE
The situation looks genuinely much better for the prospects of recovery in the FEE, which has now experienced continuing growth for a fourth year (although at a lesser rate than in 1994 or 1995). However, as noted above the current situation in Albania should warn us not to be too complacent in assuming that the main problems are over. Popular unrest is often much greater once expectations have been raised that recovery is taking place.
a) The Four Warring States
The four warring states of FEE with a population of 21.8 million cover all of former Yugoslavia except the richest state of Slovenia which has become one of the greatest success stories of transition. 1997 will undoubtedly have to include Albania in this category and its warring activities will result in great economic loss, but for 1996 Albania will remain in the category of more successful transition economies. As I noted in Part One, it was often cited to confound the pessimist.
i) Bosnia-Hercegovina (4.4 million population). Due to continued instability we still do not have any data for Bosnia-Hercegovina which has undoubtedly suffered great losses.
ii) Yugoslavia (Serbia-Montenegro) (10.5 million population) There are also no recent figures available for Yugoslavia (Serbia-Montenegro). Earlier figures had shown a dramatic fall to 47% of 1989 levels of GDP in 1993, and then a growth of 6-7% a year in 1994 and 1995, which would bring them to about 54% in 1996. No figures have been published for 1996.
iii) FYR-Macedonia (2.1 million population) The GDP of FYR-Macedonia appears to have grown by 3% in 1996 bringing a halt to the previous decline at a level reported to be 65% of 1989 levels of GDP. The per capita GDP for FYR-Macedonia in 1996 was about 832$US (Atlas method).
iv) Croatia (4.8 million population) Croatia is reported to have been increasing its GDP at a growing rate reaching a growth of 5% per year in 1996 and reportedly achieving 68-70% of its 1989 level GDP in that year. Its per capita GDP in 1996 was about 2,742$US (Atlas method) which was larger than Russia's (2,468$US) or any FSU level apart from Estonia (2,997$US).
b) The Less Developed Parts of the FEE
The 34.3 million people inhabiting Bulgaria, Albania and Romania are classified as living in the less developed parts of the FEE which are in many regards closer to the FSU levels than the more developed parts of FEE. Their average per capita GDP in 1996 was 1,281$US (Atlas method) and 4,505$US in PPP (excluding Albania) which was lower than the average FSU level of 1,865$US when measured by the Atlas method, but higher than the average FSU level of 3,289$US when measured in PPP.
i) Albania (3.2 million population) Albania will certainly need to be included as a warring state when we come to look at 1997 growth rates next year. Until Albania exploded into popular revolt, it had appeared as the most rapidly recovering country in the FEE. With its high economic liberalization index for 1995 and consistently high growth rate from 1993 to 1996, Albania appeared to demonstrate that policy was all important and that the extent of structural challenge appeared to be a secondary matter. (See references above). Albanian per capita GNP in 1996 can be estimated at about 433$US (Atlas method) which was lower than all countries in the FSU apart from Tajikistan and presumably Georgia.
ii) Bulgaria (8.4 million population) After the sharp decline of the early 1990s the apparent turn to growth in 1995 in the Bulgarian economy has been sharply reversed by a 6.5% decline in 1996. According to the IBRD figures this resulted in a 1996 level of GDP that was 70% of the 1989 level, while the EBRD claims a level of 80%. In any case Bulgaria looks set to fall out of the circle of FEE transition successes and into a position more similar to the most successful FSU countries. Given the pre-transition situation of that country and the extent to which its transition was more structurally challenging than that of most of the FEE countries, this should not be a great surprise. Bulgarian per capita GNP in 1996 was about 1,231$US (Atlas method) and 4,314$US in PPP.
iii) Romania (22.7 million population) Romania is another country that appears to be growing successfully and registered a 4.5% growth in 1996, and levels of 82% of the 1989 GDP level. However, given the recent performances in Bulgaria and Albania, I would be a little hesitant in calling this a success yet. Romanian per capita GNP in 1996 was about 1,419$US (Atlas method) and 4,569$US in PPP.
c) The More Advanced Parts of the FEE
The five countries of the more advanced parts of the FEE are by far the richest countries in the FSU and FEE. They have suffered far less economic decline than the other countries, and have in recent years all followed extensive liberalisation programmes. However, no area-study specialist is likely to claim that the main reason for these countries being more advanced is the more extensive liberalisation programmes carried out by them in the past seven years. But this is the logic behind the IMF and IBRD regressions discussed in Part One of this article. In terms of changes in the relative wealth of LDPFEE and MDPFEE it is demonstrable that the gap is widening. In 1989 the Less Developed Parts had an average per capita GNP in PPP of 5,240$US, which was 75.6% of the average per capita GNP in PPP of 6,933$US for the More Developed Parts. In 1996 the LDPFEE per capita GNP in PPP of 4,505$US was only 66.1% of the average per capita GNP in PPP of 6,818$US for the MDPFEE.
i) The Czech Republic (10.3 million population) (5% growth reaching 89% and per capita GNP in PPP of 9.766$US);
ii) Slovenia (2 million population) (3% growth in 1996 reaching 94% of 1989 level GDP and per capita GNP in PPP of 6,738$US);
iii)Hungary (10.3 million population) (1-2% growth reaching 88% of 1989 level of GNP and a per capita GNP in PPP of 6,264$US);
iv) The Slovak Republic (5.3 million population) (with a 5% growth in 1996, reaching 88% of 1989 GDP level and with a per capita GNP in PPP in 1996 of 6,264$US);
v) Poland (38.5 million population) (5% growth reaching 103% of the 1989 level GDP and with a per capita GNP in PPP in 1996 of 6,156$US).
Only with these countries can we confidently feel that we have reached an area where the Crisis Zone no longer rules and where real economic transition has occurred.
3) Conclusions
Despite the prevalent attitude of optimism exuding from the World Bank and other international agencies, I do not think that the situation has materially changed from the position of overwhelming crisis in most countries of the FSU and some of the FEE that I outlined last year.
In the five Warring States of the FSU the rate of decline (which had been enormous earlier) did abate somewhat. For the other countries of the FSU the overall rate of decline remained at about 2.6%. Considering that the FSU has now experienced seven years of economic decline which result in current (1996) GDP registering 70% (or 75% with RF corrections) of its 1989 level, the continuing failure to recover should be registering more concern than is apparent from the complacent statements of the President of the World Bank. This cannot be categorised as a 'temporary decline in welfare'; it is already the major economic crisis of the twentieth century and there are few signs that any real recovery is taking place in the economies of the FSU.
Until the recent eruption of what the World Bank will undoubtedly call 'regional tensions' in Albania, it looked as though the situation in many of the former 'Warring Zones' was improving and that the threat of 'spill over' was probably not as great as many, including myself, had feared. Albania is a sad warning for us all. History teaches that the most 'tense' period, when economic disruption is likely to have the greatest effect on popular discontent, is not when an economy is declining, but it is more frequently after a severe trough, when recovery has begun and when there has been a hold-up or temporary reversion in the recovery process. This article argues with even more vigour than last year that it is time to stop talking complacently about the imminent successful transition, and to look at the actual crisis that is upon us.
History Department
University of Melbourne
e-mail: S.Wheatcroft@history.unimelb.edu.au
Endnotes:
(1) From 1997, we will no doubt have to include Albania as a fifth Warring State in this region.
(2) The US$ equivalent of GNP for 1996 has been estimated on the basis of the US$ equivalent of GNP for 1994 as calculated by IBRD, and the growth of total GNP in constant domestic prices for 1994-1996
(3) For a discussion of these two indicators see S. G. Wheatcroft, 'Charting the Crisis Zones of Euro-Asia', Centre for Russian and Euro-Asian Studies, Melbourne University, Russian and Euro-Asian Economics Bulletin, vol. 5, no. 2, February 1996.
(4) less the four warring states, Albania, and Slovakia.
(5) Bangladesh=1,330US$ in 1994, Haiti = 930US$, and Uganda = 1,410US$ see WDR 1996, p. 188
How to Subscribe to the Russian and Euro-Asian Bulletin