BRICS and the economic decline of the old world,

Transcription

1 BRICS and the economic decline of the old world, Arne Melchior, NUPI Background note for the ONS Summit 2012: The geopolitics of energy, August This note is made available at the project web page as a background note, and the content may be used as input in other publications later. Some aspects are also summarised in the report (Eriksen et. al. 2012). Another related publication is Melchior (2012). Abstract The paper examines GDP growth and the share of different regions in the world economy , with some simple extrapolations for For several decades, there has been strong reallocation of world GDP shares from Western Europe towards. During the last decade ( ) this change accelerated, and the share of North America also fell considerably. Measured in PPP (Purchasing Power Parity, that accounts for price level differences), the share of Western Europe + North America in world GDP fell from 51% in 1990 to 42% in 2010, whereas s share increased from 23 to 34%. From the turn of the century, growth accelerated in other world regions such as Africa, Latin America, Central Europe and the Former Soviet Union. If the trend from the last decade continues, s share will rise to 57% in 2030, while North America + Western Europe will be down to 21%. This forecast may be exaggerated since China s growth is expected to slow over time, and some recent growth was stimulated by commodity price hikes. Better education and less war may nevertheless create sustained growth in the developing world and lead to a continued realocation away from the old industrial countries. The BRICS countries play an important role in this global reallocation and especially Russia, Brazil and to a growing extent China are leading in their regions. In, many nations contribute to the stronger role of the regions. India has contributed but lags behind China and the rest of should not be forgotten. In Africa, South Africa is important but a limited part of the overall story. Hence the BRICS are important but differ in importance, and they should be focused but not over-focused. 1

2 1. Introduction As shown in Melchior (2012) there has recently been a dramatic reallocation of the shares of different regions in world trade; with a rise for and a decline for Western Europe and North America. After 2000, other world regions have also expanded their shares. Corresponding to this reallocation of world trade, similar changes have occurred for shares of world GDP. This note presents an up-to-date analysis of GDP growth for , and some simple forecasts and extrapolations for The BRICS are large emerging economies representing a trend with development in new parts of the world. There are however a number of emerging economies beyond BRICS, so we are interested in developing regions and not only the BRICS. Some BRICS are large and dominating in their regions, but others represent a more modest share so that a single-minded focus on the BRICS creates a biased perception. We therefore analyse regions as well as BRICS, and examine how important the BRICS are within their regions. With respect to regions, we use the same classification as in Melchior (2012); with nine world regions: Western Europe; Central Europe (including Greece and Turkey); FSU (Former Soviet Union); Africa; Middle East (including Israel); (from Afghanistan and eastward); Oceania and Pacific; North America (including Mexico); Latin America (inncluding the Caribbean). There is also a 10th NES (not elsewhere specified) region; this is however very small for GDP data and specific results will not be reported even if the group is included in the total. 1 For details, see Melchior (2012). 2. The data set For international comparisons, GDP data could be expressed in current USD, constant USD or PPP (Purchasing Power Parity) $ - i.e. adjusted for price level differences between countries. Data in current USD can be used for calculating shares of world GDP but analysis of growth should use data in constant dollars. PPP data are based on artificial exchange rates that adjust for price levels. Since poor countries have lower price levels, PPP data give higher shares for developing countries in the world economy. Figures in constant dollars adjust for inflation in the USA but do not take price level differences into account. Figure 1 shows the number of countries covered in the data base World Development Indicators (WDI) from the World Bank, for population, GDP in constant 2000 USD and GDP in 2005 PPP$. 1 In 2010, the NES group represented less than 0.1% of world GDP. 2

3 No. countries Fig. 1: Data coverage Population 125 GDP-PPP GDP-const$ Hence data coverage is better from 1990, so we will use data for in order to increase the data coverage. In the study, we use two different data sets for : A data set in constant 2000 USD covering 180 countries and territories. A data set in 2005 PPP USD covering 172 countries and territories. In the following, we refer to these as nominal GDP and GDP-PPP, respectively. In order to obtain a complete time series for all the 172/180 countries, we use some supplementary information, especially for 2010 where some countries have not yet reported their data to the World Bank. 2 For all years, the data sets cover countries representing between 97 and 98% of the world s population. The missing countries are developing countries and territories that could be expected to have income levels below the world average, so the data sets might cover 98-99% of world GDP. When we calculate world GDP we mean the aggregate for the countries included in the respective data set. 3. Regional income levels and shares of world GDP, Appendix Tables A1-A4 show per capita income levels and shares of world GDP for the ten world regions for Tables A1-A2 are for nominal GDP, while tables A3-A4 use GDP-PPP. Figures 2 (nominal) and 3 (PPP) show the shares of world GDP for North America, Western Europe, and other regions combined. 2 We use supplementary data on GDP growth from CIA World Factbook in order to calculate GDP for in some cases. For GDP in 2000 USD, this applied to 19 countries for 2010, 5 cases for 2009 and one case for For GDP-PPP, it applied to 11 cases for 2010 and one for Furthermore, data for 1980 were extrapolated in five cases based on growth rates for the four subsequent years. These extrapolations and use of other data may represent a minor inaccuracy in the data (42 out of 7392 data entries, or 0.6%) but apply to mainly smaller countries and do therefore not affect the overall results too much. When data for are reported, the calculations may be reiterated in order to eliminate any inaccuracy. 3

4 35 Figure 2: Region shares of world GDP, (constant 2000 $) North America W Europe Others Fig. 3: Regional shares of world GDP (PPP, 2005 international $) North America W Europe Others There was a continuous decline for Western Europe and a continuous increase for, only interrupted by the n crisis in For North America and Others, the period is split in two with a reversal from 2000 (or 1999, even more accurately). As we would expect, the trends over time are similar in nominal $ and PPP, but considerably stronger with PPP data. Especially for, GDP growth is much stronger with PPP than nominal $. This is shown in Table 1 and Figure 4. 4

5 Annual growth (%) Table 1: Average annual growth rates for world regions, (%) Region Nominal PPP Nominal PPP Nominal PPP Africa Central Europe FSU Latin America Middle East North America Oceania & Pacific Western Europe World Fig. 4: Average GDP growth Nominal PPP World W. Europe Oceania & P. North Am. M. East Latin Am. FSU C. Europe Africa As shown in Fig. 4, the old world with FSU, Western Europe and North America had the lowest growth rates over the whole period. For some regions including the FSU, however, there was a sharp difference between the two decades, with lower growth in the 1990s. Figure 5 shows this difference, using the PPP data. 5

6 Fig. 5: Growth in 1990's vs. 2000's World W. Europe C. Europe FSU North Am. Oceania & P. M. East Latin Am Africa Average annual GDP growth While growth slowed down after 2000 for North America and Western Europe, it accelerated for the FSU, Central Europe, Africa, and (more modestly) Latin America. Hence a main feature is the relatively low and declining growth rates of the former economic world leaders. While the relative share of Western Europe has been declining for almost five decades, the new feature is that the USA and North America has joined the club, cutting the pace of growth by half from the 1990s to the 2000s. The long-run change is illustrated in Figure 6, based om Maddison s data set (in PPP). 3 3 Available at This data is in PPP (using the so-called Geary- Khamis approach) and covers about 200 countries (varying over time due to the formation of new countries). 6

7 % of world GDP (PPP) Fig. 6: 's growth and the decline of the West: % of world GDP Based on Maddison's data set North America Western Europe For the post-1990 period, the pattern is similar to our Figure 3 based on PPP. Figure 6 shows that the rise of and the fall of Europe extends back in time, while the USA/ North America had a stabler share of the world economy during So the new thins after 2000 is the fall of the USA, and the accelerating decline of North America and Western Europe. Observe again that we examine relative changes and that with temporary exceptions, there has generally been positive real growth in spite of the falling share for the old world. With Maddison s data, the growth of may be slightly inflated by the use of a particular method for constructing PPP data (see footnote 3). For most regions, the difference between nominal and PPP growth rates is very modest. For, however, GDP growth is dramatically higher measured in PPP$. There have been heated debates about the quality of PPP data, and their fragility is illustrated by the fact that after collecting new worldwide price data in 2005, the World Bank adjusted income level estimates for many developing countries significantly downward (for a discussion, see Deaton and Heston 2008). For example, the income levels of China and India were cut by almost 40%. There has also been critique suggesting that some methods to construct GDP-PPP were leading to an overestimation of developing country growth. According to Deaton and Heston (2008), however, the results from different metodologies are not that dramatically diffrent although some gaps exist. Hence issues related to data seem to be a more important factor creating some uncertainty. The gap between nominal and PPP growth rates for has to imply that price level changes in some way added to real growth. There was, however, a very strong price level increase in China: according to Ravaillon (2010), the price level of China doubled from 1993 to There is some uncertainty as to whether the price level 7

8 of China (and its change over time) is appropriately measured in the PPP price data (Deaton and Heston 2008). Given thart new data for previous years cannot be collected, some uncertainty will remain about PPP-based growth rates for China and. As we have seen, the qualitative trend is however the same for nominal and PPP data, although the strength of the global reallocation is stronger with PPP and this depends on price level measurement. Compared to growth and reallocation of shares in world goods trade (Melchior 2012), trends for GDP are similar but weaker. For example, the share of Western Europe in world trade fell from 46% in 1990 to 30% in This may be compared to GDP, for which the Western European share of the world total during the same period fell from 27.6 to23.1% (nominal) or from 24.5 to 18.5% (PPP). The slower decline for GDP is partly because of higher growth for services than for manufacturing, and world goods trade is strongly influenced by manufacturing activity, where the reallocation toards has been very strong. The analysis nevertheless shows beyond doubt that developing country regions grew faster during the period studied. For, this applied to both decades but after 2000, other developing regions have also accelerated GDP growth while the old world slows behind. The analysis has shown that the turn of the century was also an economic landmark; from that point the USA also fell behind, and developing regions beyond joined with varying intensity the process of growth and catch-up. 4. BRICS shares of world GDP BRIC and BRICS are symbols of this transformation of the world economy, as major emerging economies withing their regions. The growth of is however a phenomenon far beyong China and India. While China is certainly the emerging new giant, India is still at an earlier stage of development. When focusing on the BRICS, it should therefore be recalled that they differ in importance within their respective regions and that other countries also matter. Using the same definition of regions as before, table 2 shows the share of the BRICS, plus Germany and the USA, in their regions in 1990, 2000 and

9 % of world GDP Table 2: BRICS shares of GDP (PPP) in their respective regions Country Region Brazil Latin America China India South Africa Africa Russia FSU Other countries: Germany Western Europe USA North America Brazil, Russia and the USA are in increasing order dominant in their regions and China is gradually approaching a similar position in. Other countries in however also matter, as shown by Figure Fig. 7: Composition of 's growth (% of world GDP, PPP) Others India China Japan While Japan s share of world GDP has declined steadily (from 9.0 to 5.8%), China s share grew from 3.5 to 13.6% over the period. Shares of world GDP for all BRICS are shown in Table 3. In spite of fast growth recently, India is still a limited fraction of the n growth story, and other countries such as Korea, Indonesia, Thailand add up to a significant share for others. Hence the glory of India is cuirrently slightly inflated, although its important role, size and recent growth success is undisputed. 9

10 Table 3: BRICS shares of world GDP, PPP = puchasing power parity, 2005 international $ USD = constant 2000 USD Brazil China India Russia South Africa PPP USD PPP USD PPP USD PPP USD PPP USD Future predictions In the popular debate on the BRICS, numerous calculations have been made on how fast the BRICS, and in particular China, will climb on the global rankings. With respect to GDP, the former analysis has shown that the outcome of such calculations depend strongly on whether PPP or nominbal data is used. Second; the result depends strongly on the basis for extrapolations: trends have been quite different in the 2000 s compared to the 1990 s so it matters greatly what kind of growth rates that are extrapoilated. As an illustration, we can show how the world would look like in 2030 depending on whether we use growth rates from the 1990 s or the 2000 s as the basis for future extrapolation. Figures 8 and 9 show simple growth extrapolations for , using nominal or PPP data, respectively. While regional growth observed, two different paths for are shown: One path shows developments if all countries have the same annual growth rates in as they had in the 1990 s. The other path shows what happens if all countries grow as they did in the 2000 s. The corresponding figures are shown in Table 4. Whether reality will conform to any of these extrapolations is of course uncertain and we discuss below how this un certainty can be narrowed. 10

11 Share of world GDP % of world GDP Fig. 8: Regional growth extrapolations % of world GDP, from data in constant $ Solid lines: Country growth as in Dashed lines: Country growth as in North America W Europe Others Fig. 9: Regional growth forecasts , % of world GDP, from data in PPP $ Solid lines: Country growth as in Dashed lines: Country growth as in North America W Europe Others

12 Table 4: Extrapolation of GDP growth , for major world regions. Shares (%) of world GDP. Path 1: National growth rates equal to average for Path 2: National growth rates equal to average for Region Observed shares Extrapolations towards Scenario (A) Based on PPP data (PPP 2005 international $) North America W Europe Others (B) Based on nominal data (constant 2000$) Path Path Path Path Path Path Path Path Path Path North America Path Path W Europe Path Path Others Path Path The extrapolations demonstrate the dramatic nature of the trend after the turn of the century: if this continues for another 20 years, the share of Western Europe in world GDP-PPP will be cut from 19% in 2010 to 9% in 2030, and North America drops from 23 to 12%. These two world regions represented 51% of world GDP (PPP) in 1990, but would be down to 21% 40 years later. With nominal data, the relative contraction of the old world (North America + Western Europe) is a bit more modest; with a share of 35-44% in 2030 compared to 61% in 1990 and 55% in In a long-term perspective, however, even this is a significant change in the world economy. As we have seen already, s growth is higher with PPP data, but the difference between the 1990 s and 2000 s is not so dramatic. For, it therefore does not matter too much whether growth rates from the 1990 s or 2000 s are used; especially with PPP data the trend is similar in the two cases. For other regions, the pattern is the opposite: nominal and PPP trends do not differ that much, but it matters strongly whether the trend from the 1990 s or the 2000 s is applied. The difference between the two growth paths is particularly large for others, but also for North America and Western Europe, Path 2 with growth rates from the last decade leads to a faster change. 12

13 GDP-PPP These extrapolations illustrate that the recent past represents a significant rearrangement of the world economy, with new regions on the rise and stagnation for the former leaders. Which trend will apply to the next decades is however a more complicated issue, and it is beyond the scope of this paper to examine every country and provide a complete forecast. For China, the World Bank has recently (World Bank/ DRC 2012) presented growth forecasts with annual GDP growth falling from recently about 10% to 5% annually in According to the report (ibid., 9): There is broad consensus that China s growth is likely to slow, but when and at what pace is uncertain, and there is no saying whether this slowdown will be smooth or not. China s growth has been strongly investment-driven, with investment at 46.4% of GDP during (ibid.), and falling productivity growth during the last decade. Given that rsising wages may un dermine China s competitive advantage in low-skill manufacturing, future growth depends on China s ability to upgrade skills, innovate and import technology. How this will evolve over time, is hard to predict accurately. Figure 10 shows what will happen if growth continues like in (at 9.6%), compared to the World Bank/DRC forecast. For comparison we also show extrapolations for the USA; here we show both Path 1 and Path 2 since they are quite different (for China, growth in was slightly lower than during the 1990s). Fig. 10a is in PPP, 10b nominal Fig. 10a: China and USA GDP growth (PPP) China-Path 2 China-WB USA-Path 1 USA-Path

14 GDP-const. 2000$ Fig. 10b: China and USA GDP growth (2000$) China-Path 2 China-WB USA-Path 1 USA-Path With PPP data, the GDP of China will become larger than the USA level already in 2017 (Path 2) or 2019 (World Bank scenario). With nominal figures, China will surpass the USA only in the simple extrapolation scenario, but it will approach the U.S. level also in the WB scenario, where annual growth falls to 5%. China is already today the world s largest exporter and manufacturer, and the second largest economy measured by GDP. For China, an ageing population is one of the arguments for a lower growth rate in the future. For India, the demographic aspect is more favourable since labour force participation is low and may be expected to grow. For India, on the other hand, there are a other uncertainties about the future. Recent growth has been driven by services rather than manufacturing, and India resembles a dual economy with pockets of growth but large population segments left in misery. We expect Indian growth to continue, and growth rates may remain high. But in order to really succeed, India has to educate its workforce more broadly and boost the creation of jobs outside agriculture and outside the so-called informal sector that is unfortunately the industrial trade-mark of India. For every country, assessing future growth is a complex issue and it is beyond the scope of this paper to undertake such comprehensive groath forecasts. Our discussion of China illustrates that extrapolation of past growth into the future is uncertain. For Africa, growth in the 2000 s was partly driven by commodity prices and for some commodities, the trend from the last decade cannot be expected to continue over decades. For many developing regions, however, extensive investment in education in the past may provide a basis for sustainable growth in the future; in the Middle East, in Latin America and in Africa. In, many of the initial tigers grew more slowly some time after the n crisis in the late 1990 s. But some of these have solved political problems and may be prepared for another era of future growth; e.g. some ASEAN countries. With fewer armed conflicts worldwide, the developing world may be better able to obtain sustained growth. 14

15 % of world GDP-PPP With rising wages in China, new countries may also succeed in low-skill manufacturing, after three decades with tough competition from China. Taking these positive factors into acccount, it is therefore quite possible that the reshaping of the world economy may continue rapidly, even if China matures and enters into a stage with somewhat lower growth. Hence the overall conclusion is that growth in China is likely to decelerate, but growth in other developing regions may continue moer or less as in the last decade. The share of North America + Western Europe in the world economy may therefore continue to decline. In Fig.11, we have used our Path 2, with growth as in , but replaced growth for China with the World Bank forecast Fig. 11: Shares of world GDP - Path 2 with slower growth in China (extrapolations for ) NAM+WEU Others Hence even with dampened growth in China, it is possible that the reallocation between the old world and will continue. has 55% of the world s population, and its average per capita income in 2010 was at at 54% (nominal) or 62% (PPP) of the world average (6290 vs in PPP, 3343 vs in 2000$). According to simple arithmetics, would have a 55% share of world GDP if its average per capita income grows to the world average (with the current population). Growth in countries such as India, Indonesia and Bangladesh and others may contribute in this direction. In the longer run, this is likely to happen, and it will imply a continued strong reallocation in the world economy, as we have seen in the past two decades. References Deaton, A. and A. Heston, 2008, Understanding PPPs and PPP-based National Accounts, NBER Working Paper Eriksen, S. S., S. Lodgaard, A. Melchior, K. Rich, E. Wilson Rowe and O. J. Sending, 2012, BRICS, Energy and the New World Order. A study by NUPI for the ONS (Offshore Northern Seas) Summit 2012: The geopolitics of energy. NUPI Report, August 2012, available at Melchior, A., 2012, World Trade : Globalization, Regionalisation and Reallocation. NUPI Paper No. 805, August 2012, available at 15

16 Ravaillon, M., 2010, Price Levels and Economic Growth. Making Sense of the PPP Changes between ICP Rounds, World Bank Policy Research Working Paper World Bank/DRC (Development Research Center of the State Council), 2012, China 2030: Building a Modern, Harmonious, and Creative High-Income Society, available at 16

17 Appendix Table A1: Average GDP per capita in 10 world regions, in constant 2000 USD. Africa Central Europe Former Soviet Union Latin America Middle East North America Oceania & Pacific Western Europe World 17

18 Appendix Table A2: Average GDP per capita in 10 world regions, in 2005 PPP international $. Africa Central Europe Former Soviet Union Latin America Middle East North America Oceania & Pacific Western Europe World 18

19 Appendix Table A3: Shares of world GDP for 10 world regions, % based on constant 2000 USD. Africa Central Europe Former Soviet Union Latin America Middle East North America Oceania & Pacific Western Europe World 19

20 Appendix Table A4: Shares of world GDP for 10 world regions, % based on 2005 PPP international $. Africa Central Europe Former Soviet Union Latin America Middle East North America Oceania & Pacific Western Europe World 20

21 21