Version classiqueVersion mobile

How to achieve the welfare state in the twenty-first century

 | 
Roberto Kozulj

Part I. The wealth and poverty of nations in the transition from the twentieth to the twenty-first century

Chapter 3. From the breakups to the spatial reconfiguration of trade, production and consumption

Texte intégral

1We have already provided an overall analysis of some basic questions related to the concept and measurement of wealth, and to the links between urbanization and economic growth in order to clarify several aspects of the discourse on these topics in current economic literature. The aspects we will describe in this Chapter will provide a better understanding of the importance of the topic, and will reveal the impact that Asian urbanization – particularly that of China – has had in this respect.

  • 1 Or 2013/2014, when data are available.

2The description, illustrated by several graphical representations of series of data, will go as far back in time as is possible. In general, there will be data for the period 1950-2012, or for others along which there are facts that deserve observation. Three main periods will be differentiated: 1950-1975, the so-called “Golden Age”; 1975-2000, a stage with slow rate of growth and no convergence between developed and developing countries; and 2001-20121, marked by China’s accession to the World Trade Organization (WTO) in 2001, when countries accounting for 80% of world GDP had already joined it.

3China’s large urbanization process had started much earlier than that, though, along the nineteen-eighties or nineties, as a consequence of economic reforms and the fact that the most restrictive stages of rural-urban migration were already over. This was also the time of preparation for “the great leap forward”, or what some people consider the Communist Party’s decision to implement the model called “caged capitalism”, where the bird represents capitalism and the cage, the State. The choice of these three stages aims to show crucial differences in world economic dynamics in relation to urbanization processes and their several impacts along their development, according to the arguments presented so far, and also as the basis for an analysis of future challenges.

  • 2 Basically, all those institutions belonging to the UN system, such as the IMF, the WB, regional bod (...)

4The world economy has been growing relentlessly for sixty two years or so, creating the attractive – albeit dangerous – illusion that social progress is universal and that, sooner or later, will include all human beings. Emphasis on whether the driving force is the market and the institutions created along the second half of the twentieth century2, or growing State intervention – whether or not including ownership of the means of production – has fluctuated considerably. So have theoretical and conceptual approaches, discourse and pragmatic combinations.

5These discussions have also permeated the economic theory and have led to a revision of the history of economic thought and of the moment in which the neoclassical and neo-Keynesian synthesis gave rise to the divide between micro and macroeconomics. There has been renewed interest in Leon Walras’s initial suggestions, which has led to the finding that, in the roots of the literature on socialist planning, typical of the nineteen fifties and sixties, authors such as Oscar Lange also derived their rationale from Walrus.

6After each capitalist crisis, authors such as Eric Hobsbawm have suggested that, in the end, Marx’s approach is the only one able to predict the great final crisis of the system. Indeed, in his introduction to the Manifesto, Marx says:

The bourgeoisie cannot exist without constantly revolutionizing the instruments of production. […] The bourgeoisie has through its exploitation of the world market given a cosmopolitan character to production and consumption in every country. To the great chagrin of Reactionists, it has drawn from under the feet of industry the national ground on which it stood. All old established national industries have been destroyed or are daily being destroyed. They are dislodged by new industries, whose introduction becomes a life and death question for all civilized nations, by industries that no longer work up indigenous raw material, but raw material drawn from the remotest zones; industries whose products are consumed, not only at home, but in every quarter of the globe. In place of the old wants, satisfied by the production of the country, we find new wants, requiring for their satisfaction the products of distant lands and climes. […] And as in material, so also in intellectual production. […] The bourgeoisie has subjected the country to the rule of the towns. It has created enormous cities, has greatly increased the urban population as compared with the rural, and has thus rescued a considerable part of the population from the idiocy of rural life. Just as it has made the country dependent on the towns, so it has made barbarian and semi-barbarian countries dependent on the civilized ones, nations of peasants on nations of bourgeois, the East on the West” (Marx & Engels, 1848).

  • 3 Cf. Zizek, S. (2002, p. 22).

7Then, intellectuals wonder in amazement: Is this not, more than ever, our present situation?3 Now empirical evidence reveals that these ideological discussions – related to the nature of the ecology of the ideas – can hardly contribute a solution to the problems humanity is facing and will continue to face unless the debate is rich in new ideas, approaches and conceptual and theoretical frameworks.

8The warning here has to do with the fact that the limits to expanded accumulation and reproduction in world economy may be based on factors that have not been sufficiently stressed in the current views on the topic, and which are related to structural aspects of a technological nature in turn conditioning key economic factors such as employment and profitability. An already quoted example is the link between stages of urbanization with certain demand structures associated to them, and a strict international division of tasks in a world dominated by invention and patents.

9Such limit could obviously derive from the lack of “new things to do”, – when most of the cities are already built – because of the lack of strategic planning in the field of economic supply, and of tools aiming to widen the demand for goods and services that, for several reasons, some people cannot afford to get. Among those reasons is the decline in certain activities alongside the saturation of urbanization processes.

10This, of course, introduces a serious problem regarding equality in a wider sense than is normally implied. It is not just a question of improving income distribution, but of not generating a gap between those who have an income coming from their jobs and those who have it as a universal right, whether they work for it or not. What is crucial here is the discussion about the limits of meritocracy and its likely legitimacy as a criterion for distribution. For, though certainly structural unemployment is a cause of the lack of remuneration, having a universal right to a salary might deepen the social gap and hurt the relation between “effort” and “remuneration” probably to a point of no return.

11This could be illustrated by the youth revolts in Paris, London and other cities, where the pattern of the looting in which they were involved reveals that this was not the consequence of hunger but of a belief that could be translated as follows: “I also have the right to an iPhone, or a smart TV, and society does not provide me with legitimate means of getting them”. This is, of course, an ideological distortion of social struggles, but what can we say to the young people regarding their right to have a house or to feed their children? That this is not a subjective exaggeration or a tendency to criticize the system is indicated by some specific and well grounded research (cf. Bisang & Campi, 2010).

12On the contrary, can society ignore that its efforts to create employment will fail if they are not conceived as segmented by categories of human capital and type of activity? Can technological innovation applied on already established uses generate such mass of employment? Can new large technological revolutions do it? All these questions have so far no more empirical bases than those revealed by data in retrospect and, according to them, answers are more negative than positive, though not categorical.

13From the point of view of what these issues imply for the “more market or more State" dichotomy, the data do not contribute much either. Twentieth and twenty-first century evidence so far reveals more cases of socialist and communist economies searching for market mechanisms to promote development. However, crises such as those in capitalist countries that have tried market reforms – and particularly the crises in central countries – suggest that problems cannot be solved by State intervention, or only with certain targeted public policies. A return to socialism, in turn, poses more complex issues. It would then seem healthy to base our understanding of reality on facts, represented by means of performance indicators.

14The following Figure shows an interesting phenomenon in this respect, for it reveals the impact of Asian development and urbanization on the total growth of developing countries as a creation of gross flows of annual wealth, according to Maddison’s data series of gross domestic product. It should be noted that UNCTAD 1970-2013 series – expressed in current values according to exchange rates in each country – show that only by 2013 would the rest of the world product equal that of the fourteen most developed countries.

Figure 5. Growth of World Gross Product: developed and rest of the world between 1950 and 2008

Figure 5.  Growth of World Gross Product: developed and rest of the world between 1950 and 2008

Note: Estimated on the basis of Maddison Historical GDP Data. Values expressed in “Geary-Khamis dollar” (GK$), a hypothetical unit of currency that has the same purchasing power parity that the U.S. dollar had at a given point in time – in this case, 1990. Comparison of this long series with UNCTAD 1970-2008 data behaves in a similar – though not identical – manner.

15The prevailing thought in the nineteen sixties suggesting a purported convergence of levels of wealth has been amply debated upon since then. For instance, Barro & Sala-i-Martin’s well-known analysis (1992) confirmed such convergence within the United States; Rodrick’s more recent work (2011) was along similar lines, whereas Quah’s (1995) actually questioned the former. The point is that the data expressed as average of flows of created annual wealth representing aggregates of goods and services showed slight evidence of convergence, even without considering the levels per individual.

16On the other hand, another one of Piketty’s works (1997) questions the convergence between rich and poor countries. He mentions factors such as prior inequality structures, differences in human capital, participation – or lack of it – in the world market and capital market failures to explain failures in the purported model.

17Thus, until the emergence of the China (and other Asian countries), an increase in divergence is observed, particularly between 1980 and 2000. Yet as of 2003 to date, the process has been reversed, but still without reaching the expected convergence in levels of wealth per individual. The China effect thus impacted on world economy, but in particular on developing countries, which along the first and a half decade in the twenty-first century suggested a certain illusion of convergence, anyway still far from consolidated.

Figure 6a. Growth in Gross World Product per capita: developed countries and rest of the world between 1950 and 2008 (equivalent to dollars of 1990)

Figure 6a. Growth in Gross World Product per capita: developed countries and rest of the world between 1950 and 2008 (equivalent to dollars of 1990)

Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows the trend in GDP per capita in GK$ per capita The right axis shows the value of the coefficient between such GDP per capita in the most developed nations with respect to the same average indicator in the rest of the world.

Figure 6b. Growth in Gross World Product per capita: developed countries and rest of the world between 1970 and 2013 (in constant dollars of 2005)

Figure 6b. Growth in Gross World Product per capita: developed countries and rest of the world between 1970 and 2013 (in constant dollars of 2005)

Note: Estimated on the basis of UNCTAD data of GDP at constant 2005 values, updated 1970-2012 population series, and 2009-2013 rates according to United Nations projections. The left axis shows the trend in GDP per capita in dollars of 2005 per capita. The right axis shows the value of the coefficient between such GDP per capita in the most developed nations with respect to the same average indicator in the rest of the world.

  • 4 Austria, Belgium, Denmark, Finland, France, Germany, Italy, The Netherlands, Norway, Sweden, Switze (...)

18As shown in Figures 6a and 6b, GDP per capita growth trends clearly reveal that not even the development deriving from China’s accession to the WTO has contributed to shortening the gap in wealth creation per capita between the rest of the world and the twelve richest countries in Europe4, the United States and Japan, or between the rest of the world and the set of developed nations as defined by the UNCTAD. In fact, the ratio of GDP per capita between developed countries and the rest was 4.9 in 1950 and in 2008, but reached a peak value of 6.6 in 1992.

19These values are still more drastic when expressed in constant dollars of 2005 (Figure 6b, UNCTAD data) for the ratio of inequality in annual wealth creation per capita, which was 22.5 in 1970, only slightly decreased by 20.5 in 1982, and reached a value of 22.8 to 20.3 between that year and 2004. And it decreased again by 14.9 only in 2012, after the rest of the world felt the impact of the China effect or of the two-speed pace of world growth. Such impact was felt on quantities, but also on absolute and relative prices because, for the first time since the late nineteen seventies, exchange terms improved again in favor of countries exporting raw materials, energy and food.

20This becomes more forceful when considering Figure 7a, which illustrates the annual changes in created wealth expressed in flow per capita (GDP per capita) in the developed world. This includes the twelve richest countries in Europe, Japan and the United States (all of them accounting for 20% of world population in 1950 and now for only 12%). These values are contrasted with wealth created by the economic systems accounting today for almost 90% of world population.

21Thus between 1950 and 1990, the annual wealth created in central countries was each year – and on average – almost tenfold larger than wealth created per individual in the rest of the world. However, this figure decreased between 1990 and 2008 when, on average, wealth created per individual in developed countries only accounted for 125% in excess of that created in the rest of the world.

22Just as was done with total GDP, so these figures can also be estimated using UNCTAD data, but only for the period 1970-2013, and the results are still more drastic. Between 1970 and 1980, incremental creation of wealth per capita was, on average, tenfold higher in the fourteen richest countries than in the rest of the world, but this ratio escalated to 47 on average between 1980 and 2000, with two inverse peaks. The first one was recorded in 1986, the year of the oil counter-shock which marked the end of the Iraq-Iran conflict, when terms of trade plummeted almost all around the developing world, and when impacts of the international financial system were felt on countries with great indebtedness. The second one was in 1999, the year of the recessive trend in the developed world, whose industry, as will be seen, increased alongside de-industrialization in most of the developing world. Then, with the China effect, this ratio was 7 between 2000 and 2013, and almost 0 towards 2013. This result, however, is more a consequence of the recession affecting rich countries than of the creation of average wealth in the developing world (cf. Figure 7b).

Figure 7a. Growth of incremental Gross Domestic Product per capita: developed countries and rest of the world between 1950 and 2008 (values equivalent to dollars of 1990)

Figure 7a. Growth of incremental Gross Domestic Product per capita: developed countries and rest of the world between 1950 and 2008 (values equivalent to dollars of 1990)

Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows the annual variation of GDP per capita in GK$ per capita (lines). The right axis shows the value of the coefficient between such variation in annual GDP per capita in the most developed nations and the same average indicator in the rest of the world (bars).

Figure 7b. Growth of Incremental Gross World Product per capita: developed countries and rest of the world between 1970 and 2013 (in current dollars)

Figure 7b. Growth of Incremental Gross World Product per capita: developed countries and rest of the world between 1970 and 2013 (in current dollars)

Note: Estimated on the basis of UNCTAD data, according to GDP at constant 2005 values and 1970-2009 updated population series, and 2009-2013 rates using United Nations projections. The left axis shows the annual variation of GDP per capita in dollars of 2005 per capita (bars and dark line). The right axis shows the value of the quotient between such annual incremental GDP per capita in the most developed nations and the same average indicator in the rest of the world (dashes).

23No doubt the change in average wealth creation per capita between rich countries and the rest of the world was the consequence of the emergence of China as the factory of the world, a process that enabled developed countries not to reduce but even to increase their industrial system until 2006-2007. This will be dealt with below for the data show, on the one hand, several related aspects prior to the 2009 great financial crisis, and on the other, what it meant for rich countries to expand their industrial systems outside their borders, a process that had a highly positive impact until 2006/2007, and a negative one after 2007/2008.

24A curious aspect revealed by these data has to do with the fact that the incremental maximum of physical wealth creation per capita (that expressed by GDP in constant value) in the developed world could not rise above such levels even in the most prosperous years, as was the case in 1984-1988 or 2000. Yet it had been growing until 1973 with almost no signs of destruction of the flow of wealth, as in 1958. On the contrary, there are recurrent crises and slow-down periods, and this implies that it was difficult to keep that average wealth growing (cf. Fig. 8). This is no doubt related to how that wealth was distributed, for if the incremental average could not be sustained, it is most likely that some social groups will have seen their capacity to generate and access wealth decrease, while others will have increased it.

25Besides, when the same phenomenon is analyzed in constant dollars of 2005, wealth creation levels between 1981 and 2002 appear to be higher than those between 1970 and 1980, and more unequal with respect to developing countries. This reveals the unequal impact of inflation as of the nineteen eighties, when it also reached the value of assets, and how effective it was for developed nations to formulate policies meant to restrict development in the rest of the world. It was again evident in that decade that the international financial system was dominated by developed countries, which seemed to be the case again towards 2014-2015.

Figure 8. Trend in product per capita between 1950 and the 2009 crisis: values in the series with annual variations of GDP per capita in three long periods

Figure 8. Trend in product per capita between 1950 and the 2009 crisis: values in the series with annual variations of GDP per capita in three long periods

Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows the trend in GDP per capita in GK$ per capita. The right axis shows the value of the annual increase in GDP per capita expressed in the same unit.

26The most interesting aspect here has to do with what happened with developing countries over the last years of the first decade in the twenty-first century. Over the second half of the twentieth century, the rest of the world created an average incremental wealth per capita of 50 dollars of 1990, whereas in the developed world that average was some 350 dollars. In turn, in the first decade of the twenty-first century, the developing world reached an average of 200 dollars, a fourfold rise with respect to the previous fifty years, and developed countries became stable at 356, barely the same as over the previous decades. These figures certainly have to do with the impact of urbanization. And they also represent the expansion to new regions of the same system of production and consumption, from the point of view of its technology and productive structures. And this deserves consideration.

27For instance, along the period 1950-2012, wealth creation per capita on a world scale remained fixed when measured as the quotient between global GDP increase by five-year periods and world population increase by the same number of years. But this variable decreased with respect to population in cities with over 750 thousand inhabitants and also to total urban population (cf. Fig. 9).

Figure 9. Increase in wealth creation and in total and urban population and population in cities with over 750 thousand inhabitants. Values in thousands of individuals, in GK$ and in GK$ per every thousand individuals by five-year periods

Figure 9. Increase in wealth creation and in total and urban population and population in cities with over 750 thousand inhabitants. Values in thousands of individuals, in GK$ and in GK$ per every thousand individuals by five-year periods

Note: Estimated on the basis of United Nations 2012 data, Maddison Historical GDP Data, UNCTAD, UNCTADSTAT, to connect GDP growth by 2012. The left axis shows the values of the five-year period increase in total, urban and large cities population in thousands of people (bars). The right axis shows the value of the quotients between the five-year period increase in GDP and the increase in population according to the same classification in 1990 GK$ values.

28Convergence between the rise in wealth created by urban resident and by resident along the series clearly shows that it is urban population that increased over the last five-year periods (cf. Fig. 10). This reveals a trend towards saturation of the rural-urban migration process and a decline in total population growth rates. No doubt this phenomenon poses new challenges for, as was discussed in the previous Chapter, there is a projected future stagnation of rural population growth and an increase in urban population in excess of, or equal to, the increase in total population.

Figure 10. Population increase in cities with over 750 thousand inhabitants and in urban population with respect to total population increase

Figure 10. Population increase in cities with over 750 thousand inhabitants and in urban population with respect to total population increase

29This is no trivial issue given the key hypothesis in this book: that urbanization is one of the main drivers of economic growth in a very different sense from what is suggested by other explanations – though not denying them, as was analyzed in Chapter 2. And such difference lies in the fact that activities properly belonging to the construction of cities and urban systems are embodied in GDP (or flow of annual wealth). These activities are related, as has been said, to industries such as cement, steel, aluminum, machinery and capital goods for infrastructure and setting up of other industries, whose peak would depend on prior rural-urban migration processes. These processes, in turn, would determine peaks in installed capacities and fluctuations in the use of those capacities that are not short-termed, and because of this, they may imply the destruction of those capacities. This is why creative destruction is considered a driver of growth. But the question, as has been suggested, is whether the amount of output from these processes is or will be higher or lower than the amount that is destroyed, and who is to benefit from it. If it is higher, and can absorb human capital able to adapt in terms of knowledge and abilities to those new things to do, it will be a minor problem, though requiring efforts to promote such adaptation. But if it is lower, and besides implies processes of automation and formation of human resources with abilities requiring sophisticated knowledge, then the question of exclusion as a structural phenomenon may reach huge proportions.

30However, rural-urban migration processes are of a different nature, and they are driven by other factors, which makes it difficult to understand the phenomenon. China, for instance, fully emerged as the factory of the world only after its accession to the World Trade Organization in 2001, when countries accounting for over 80% of world GDP had already signed free trade agreements. Yet the maximum incremental growth in China’s cities was recorded earlier, between 1995 and 2000 (Figure 11).

31Prior to this growth, the largest contribution of flows of foreign capital as a percentage of GDP had been recorded in China (between 1992 and 1998, according to UNCTAD), and such flow has never stopped, though it declined only slightly in 2009 (Figure 12). The accumulated stock of this foreign investment between 1970 and 2013 – without applying depreciation – accounts for only 15% of GDP in 2013. Thus, China’s trade opening and investment policies have undoubtedly been a driver of urbanization and modernization, but in turn, urbanization, as will be seen, has been as significant an engine of domestic growth as exports, if not more so.

Figure 11. Population increase in cities with over 750 thousand inhabitants and in urban population in continental China. Historical data for 1950-2010 and projections by 2025 in thousands of individuals

Figure 11. Population increase in cities with over 750 thousand inhabitants and in urban population in continental China. Historical data for 1950-2010 and projections by 2025 in thousands of individuals

Note: Estimated on the basis of United Nations 2012 data. The left axis shows population values by urban and large cities population (bars). The right axis shows the values of the increase in such population by five-year periods.

Figure 12. Trend in foreign investment in China in annual million dollars, as a percentage of GDP and as accumulated stock without applying depreciation

Figure 12. Trend in foreign investment in China in annual million dollars, as a percentage of GDP and as accumulated stock without applying depreciation

Note: Estimated on the basis of UNCTAD 2014 data. The left axis shows the percentage accounted for by accumulated foreign investment received by China in relation to GDP. The right axis shows annual and accumulated investments without depreciation.

32Large cities in China, then, grew by over 100 million inhabitants in only one decade (1990-2000), whereas along the four preceding decades (1950-1990) they grew by about 83.8 million. Between 2000 and 2010, when all of China’s export potential was at its peak after it became a WTO member, its large cities grew by 93.5 million, that is, at a lower rate than over the prior decade. In turn, total urban population growth was 237 million between 1950 and 1990, and 357.5 million between 1990 and 2010, without recording any peaks between 1950 and 2010, though the peak of population increase in large cities had been recorded in 2000.

33The purpose of this detailed report is twofold. Firstly, we mean to show that growing urbanization in China was prior to its accession to the WTO and is related to the country’s political reforms over the nineteen eighties. Yet this analysis does not overlook the fact that between 1980 and 1990, exports grew in current values at average rates almost 3.7 times higher than those of GDP (and in 1990-2000, only 1.8 times higher, with GDP growth rates in that decade three times higher than the average in the previous one). Secondly, at the time when China became a WTO member, relative urban growth was higher in intermediate than in large cities.

34In fact, between 2000 and 2010, almost all the largest cities - Shanghai, Shenzhen, Chongqing, Guangzhou, Guangdong, Wuhan, Foshan, Shantou, Tianjin and others – recorded a much lower population increase than between 1990 and 2000. Yet, they accounted for 30% of the growth recorded in large cities over that decade, and were some of the most important harbor hubs, with a key role in the following decade given the export-oriented nature of China’s economy. Beijing, on the other hand, grew more between 2000 and 2010.

35China’s economic growth increased when it became the factory of the world, and its exports of goods outstripped those of the United States as of 2007, whereas exports of goods and services only did it in 2012 (Figure 13). It is then possible to infer that the country’s high urbanization rates preceded and have been important drivers of Chinese growth, accounting, in turn, for its role in the new world economy.

Figure 13. Trend in total exports by China and the United States of America. Products, and Goods and Services in millions of current dollars between 1980 and 2012

Figure 13. Trend in total exports by China and the United States of America. Products, and Goods and Services in millions of current dollars between 1980 and 2012

Note: Estimated on the basis of UNCTAD data.

36That this process was boosted by foreign investment does not interfere with the explanations suggested here, particularly because it is crucial to the question of the creation of wealth as flow and as stock and to the problems deriving from this. That is, our premise is that some activities tend to come to an end and are not replaced with others with equivalent values, and that this is not correctly perceived. Because of this, suitable reconversion activities should be planned for the available human capital to carry out while, at the same time, such labor force becomes qualified for other tasks requiring higher levels of education and training.

37China’s growth, then, can be said to have been mainly based on the construction of a huge urban and inter-urban infrastructure. That this was so in order to promote a growth model based on exports of manufactured goods to the rest of the world does not mean that the contribution of these exports to GDP was larger than that of the activities involved in the construction of that export model. In fact, according to UNCTAD data, world total exports accounted for an average 15% of nominal GDP between 1980 and 1995 and reached a maximum 26% in 2008, and 25% in 2013. Only in 2009 it decreased by 21%, but then it remained stable at around 24-25%. In China, exports only accounted for 5.6% of current GDP in 1980, and 17% in 1995. But as of 2003, when it was 26%, it started to grow till it reached almost 35% in 2007. As of the 2009 crisis, it was around 25%, almost like the world average.

38So if the question is whether China’s exports were its main driver of growth, the answer should be affirmative but ambiguous. Indeed, it is necessary to consider in turn that this export process meant a fast urbanization process that was prior to it, and was the consequence of changes in domestic policies meant to permit controlled rural-urban migration and to get ready to become the factory of the world and fully incorporate to modernization. Besides, both processes (export and urbanization) had impacts on one another. Over the last years, however, some have been wondering whether China will not get caught in the so called “middle-income trap”, a debate that is not yet over, as can be seen in Sharon Kahn’s contribution (2012), the World Bank report on China 2030, and the joint work by Zhang, Yi Luo, Liu & Rozelle (2013). Of course we cannot judge on this debate, but we can warn that, behind this purported trap, there is the fact that the country’s urbanization has been a component of its growth, and that, inevitably, this will not continue forever, just as has happened in many other countries.

39In turn, the impact on other regions was the result of the imports that China needed – and still does. This is why it is important to consider the differential behavior of Chinese imports according to the goods imported. Then, the country’s import rates by major items have varied along the different urbanization stages. For instance, steel imports started to decrease in their dynamics in 2003 – and peaks of population in large cities were recorded towards 1995-2000 –, machinery and equipment imports, between 2004 and 2006, other goods in 2007, but food imports decreased only during the 2009 crisis (Figure 14). It can be assumed that the bulk of China’s machinery and equipment was imported from developed countries as an essential part of the foreign investment process of those countries in China, which permitted the building of the factory of the world and its associated infrastructure. If this is so, the variation in the Industry Value Added rate in developed countries may be compared – by way of hypothesis – to variations in the rate of machinery and equipment imports (Figure 15).

Figure 14. Trend in annual variations of China’s imports between 1995 and 2009 (value of the base 100 index in 2000)

Figure 14. Trend in annual variations of China’s imports between 1995 and 2009 (value of the base 100 index in 2000)

Note: Estimated on the basis of UNCTAD data.

Figure 15. Trend in annual variations in China’s imports of machinery and equipment between 1995 and 2009 and in Industry Value Added in developed countries (value of the base 100 index in 2000)

Figure 15. Trend in annual variations in China’s imports of machinery and equipment between 1995 and 2009 and in Industry Value Added in developed countries (value of the base 100 index in 2000)

Note: Estimated on the basis of UNCTAD data.

40An interpretation of the numerical data of Figure 15 above as from an extremely simple econometric model results in the data in Table 1 below. These data are quite satisfactory considering both the value of the regression by the least squares method, and the value indicating a lack of autocorrelation in the series and the significance of the corresponding t-tests.

Table 1. Results of the regression of the annual variation in the Industry Value Added index in developed countries with respect to the variation in the index of machinery and equipment imports demanded by China

Dependent Variable: INCREMENTAL_INDEX_IVA__

Method: Least Squares

Date: 10/07/14 Time: 18:48

Sample: 1996 2009

Included observations: 14

Variable

Coefficient

Standard Deviation

t-Statistic

Probability

INCREMENTAL_IMP_MACH_CHI

0.249433

0.031337

7.959800

0.0000

C

-4.486919

1.271076

-3.530015

0.0041

R-squared

0.840761

Mean of the dependent variable

2.406413

Adjusted R-squared

0.827491

Standard deviation of the dependent variable

8.381636

Standard Deviation of the equation

3.481245

Akaike info criterion

5.464221

Sum of the squared residuals

145.4288

Schwarz criterion

5.555515

Log probability

-36.24955

Hannan-Quinn criterion

5.455770

F-statistic

63.35841

Durbin-Watson statistic

1.922510

Prob(F-statistic)

0.000004

Note: Estimated on the basis of UNCTADT data. The E-Views model was used. INCREMENTAL_INDEX_IVA = inter-annual differences in the Industry Value Added in developed countries. INCREMENTAL_IMPO_MAQ_CHI = inter-annual differences in the value of Chinese machinery and equipment imports.

41It is true that such simple data could probably not explain so complex phenomena. However, there are several elements that could account for the reasons why China’s growth has been functional both to reactivating the industry in developed countries and to understanding their post 2007-2008 long-standing crisis, which is not related only to the financial crisis, as current literature has led us to believe. The correlation has to do with variations in the level of machinery and equipment concentrated in developed countries and imported by China, and with variations in industrial output in those countries, two clearly independent series when regarded as sources of information.

42We will now examine again, then, the change in the global dynamics of world economy along three periods: 1950-1977, the Golden Age (though loosely considering the date it ended); 1978-2000, the empowerment of the financial system and moderate global performance; and as of the first decade of the twenty-first century, when the China effect is relevant (cf. Figure 16) and, therefore, also urbanization as the cause of activities embodied in GDP. This is, precisely, what we mean to remark here given the importance of the topic with respect to the creation of wealth as stock and as flow. This underlies Piketty’s proposals, so topical nowadays, but also the literature on development, convergence, limits to growth and accumulation, unemployment, and other related issues.

Figure 16. Trend in global growth rates of world economy between 1950 and 2012

Figure 16. Trend in global growth rates of world economy between 1950 and 2012

Note: Estimated on the basis of Maddison Historical GDP Data and UNCTAD, UNCTADSTAT, to connect GDP growth by 2012. The left axis shows world GDP in GK$. The right axis shows the percentage of the rate of variation in GDP year by year.

  • 5 When this part of the book was written, certain crises could not yet be foreseen in all their crude (...)
  • 6 UNCTAD Stat, “GDP by type of expenditure and Value Added by kind of economic activity, annual, 1970 (...)

43It is clear that transferring industrial activities to China, busy in its modernization process and its accession to the WTO, reactivated the global economy in such a way that the phenomenon of “a two-speed economy” was only evident after the crisis. Indeed, it is as of 2008 that a certain geopolitical question has also been stressed, which will determine a growing trend to date5. When considering the annual variation in Industry Value Added (IVA) between 1970 and 2011 according to data expressed in UNCTAD current values6, several interesting conclusions can be drawn. In the first place, until the end of the Golden Age, and even until the nineteen eighties, the IVA in developing countries grew annually by amounts that, though lower than those of developed countries, were significant and positive. Besides, between 1978 and 2000, the incremental IVA in developing countries was, for long periods, zero and even lower with few exceptions (as in 1995, for instance), though the incremental IVA in developed countries was considerable, but fluctuating and accounting for almost 100% of the variation in world IVA. Finally, after 2000 – in 2003, to be precise – there is an almost equivalent growth of IVA in developed and developing countries, though since 2004 developing countries have permanently outstripped developed ones, and since 2007 the IVA in developed countries has fallen or stagnated (Figure 17).

  • 7 This explanation is neither untimely nor ad hoc, as a review of the article in question will reveal (...)

44Summing up then, as shown, to some extent, by the correlation between the variation in China’s imports of machinery and equipment and the annual variation in IVA in developed countries (Figure 15, Table 1), the cause of such behavior was not China’s slowdown in exports but, most probably, the saturation of installed capacity to build the factory of the world, implied in the country’s urbanization peaks towards 2000. Now this was to be expected, according to previous work (Kozulj, 2003) written when China was not yet considered the factory of the world7.

45The article referred to was written between 2000 and 2001 on the basis of studies carried out in the mid nineteen nineties, and it anticipated a likely overcapacity crisis in 2006-2009 by means of the decomposition of a population series, both historical and projected, in large cities into two Bell curves from a logistic simulation using United Nations data. The analysis also showed that those peaks would take place five to ten years before the market saturation deriving from the building of urban lifestyles. In the case of the end of the Golden Age, peaks in population increase in large cities took place towards 1960-1965.

  • 8 Heteroskedasticity is usually produced when working with cross-sectional series containing samples (...)

46My emphasis on the fact that peaks of urban population increase were prior to product growth may be interpreted as though I wanted to point out some type of “causality”. Yet this is not so, strictly speaking, and it would not be easy to show that with econometric methods, though they were used. As was the case in that work mentioned above, I only want to remark that the correlations between urban population increase and in cities with respect to the out-of-phase increase in GDP produce better results than when synchronic models are used, or when the growth in urban population is explained through the subsequent and/or prior increase in GDP. Obviously, as will be seen below, the data available to carry out such correlations can hardly avoid the problem of heteroskedasticity8.

Figure 17. Trend in annual variations in Industry Value Added (IVA) in developed and developing countries and on a world scale (1970-2011). In millions of current US dollars

Figure 17. Trend in annual variations in Industry Value Added (IVA) in developed and developing countries and on a world scale (1970-2011). In millions of current US dollars

Note: Estimated on the basis of UNCTADSTAT data. Chosen variable: annual changes in the absolute value of Industrial Value Added (IVA) at current prices and according to current exchange rates in the countries in question.

47What has been said so far stresses an infrequent approach regarding the causes of deep structural crises related to urbanization processes. The cases of China and other Asian countries introduce new complexities in this respect. China’s urbanization has been – and still is – an important driver of industrial reactivation in developed countries. In turn, transferring industrial activity to China has posed new challenges to developed countries, for their institutions seem not to be completely ready for that and their own capacity to generate employment has been upset as a result. Besides, the international division of technological and industrial specialization has become more marked with developed countries leading such technological innovation in all sectors. On the other hand, relocating industrial activities outside their boundaries implies several inter-dependencies for developed countries, particularly regarding security and a new geopolitics due to two reasons: China needs to secure its supply of raw materials, food and energy, and it might – for the limits are not clear in this respect – quickly become an innovator/adaptor. This might generate serious problems for the continuity of the creation of flows of wealth in many developed nations that can still guarantee inclusion of their citizens regarding what they had always offered as an incentive (consumption levels depending on flows of income of created annual wealth).

48This will be dealt with again below, but it is necessary to emphasize here that the benefits accruing from trade with China cannot be considered only on the basis of the balance of flows of goods. Another advantage in this respect has to do with the fact that such trade has led to mercantilist accumulation since it has captured a vast trade income as a consequence of the differences between the cost of production of Chinese products and their sale prices in countries with high income levels. This may probably explain, at least partly, why the largest contribution to the creation of new wealth in developed countries came from the service sector, apart from the impact derived from telecommunications industries and their complex value chains.

49Then, in order to complete the description of the impacts that can be represented by indicators, it is worth mentioning that out of the incremental total of created wealth in the most changing part of the twenty first century (that is, 2003-2011), 44% was generated in developed countries and 56% in developing ones. Yet the sectoral contribution to the new wealth by type of activities was completely uneven. Developing countries accounted for 90% of the total increase in agricultural production, 69% of industrial production increase, and only 46% of increase in the service sector, for an 89% increase in world urban population between 2000 and 2010. In developed countries, on the other hand, the service sector contributed 76% of the increase in new wealth as annual flow, whereas industry and agriculture contributed only 24%. All this is shown in Figure 18 below.

Figure 18. Trend in accumulated variations in value added by large sectors in developed and developing countries. Data shown as the difference between 2003-2011 in millions of current US dollars

Figure 18. Trend in accumulated variations in value added by large sectors in developed and developing countries. Data shown as the difference between 2003-2011 in millions of current US dollars

Note: Estimated on the basis of UNCTADSTAT data, 1970-2011.

50Then, the Industry Value Added in developed countries, which, according to UNCTAD data at current prices, represented 31.7% of total world IVA (and which reached a minimum of 24.1% in 1988), accounted for 50% in 2011. That is, there was a real structural change regarding the spatial configuration of production, trade and consumption, particularly because for the first time since the peak in industry and capitalism, there was a record share of the South with respect to the North. Whether this implies a real change in power relations is a different issue, which will be discussed below.

51However, the maximum share of developing countries prior to China’s becoming the factory of the world was 35% towards the end of the nineteen eighties, concomitant with the success of Asian southeastern countries. After that period, particularly between 1990 and 2000, the share of developing countries in the generation of world IVA fell alongside the so called “Great Moderation” period, an expression that describes the reduction in volatility of American economy (Stock, 2002; Stock & Watson, 2003). Then one might wonder whether the control of business cycles in developed countries under the absolute leadership of American policies had nothing to do with the low industrial performance of the rest of the world until before the China effect.

52Between 1990 and 1999, the share of developing countries (almost 88% of the world population) in IVA was only around 25%, with 5% variability. But the curious thing is that this industrial stagnation in developing countries did not imply the absence of deep crises in the United States even in the years of the Great Moderation. The Table below Figure 19 shows a decrease in the variability coefficient of the rate of growth which is far from a control of the cycles for an economy in a stage of moderate growth.

53After 2008 a vast amount of literature tried to explain the financial crisis that took place in that year by attributing it to aspects of the monetary policy and of financial regulation. However, some remarkable studies have pointed to the fact that such volatility had already shown its severity back in 2001, and they explain why not all the emphasis should be placed on monetary and fiscal policies considering the context of the real American economy between 2001 and 2004 (Adam & Vines, 2009).

54Introducing a brief digression into the present line of reasoning, Figure 19 below illustrates how the Great Moderation period recorded three instances of deep activity decline in the last two decades, even when it does not show what happened after 2008. On the other hand, no doubt the policies implemented also resulted in real activity, particularly related to house building.

Figure 19. Trend in GDP in the United States and in the fluctuation of inter-annual rates of growth in 1916-2008

Figure 19. Trend in GDP in the United States and in the fluctuation of inter-annual rates of growth in 1916-2008

Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows GDP values and the right axis, those of its inter-annual variation.

55Still as part of the digression introduced above, it is possible to illustrate the behavior of Chinese economy in its world ascent between 1980 and 2008, when the percentages of variability of GDP growth were curiously lower than those in American economy, according to the same data source for 1950-2008 (Figure 20). Towards this last year, the total size of Chinese economy reached 93% that of American economy expressed in dollars and according to an equivalent purchasing power parity. Yet, it was 57% expressed in current dollars and only 35% if estimated at constant dollars of 2005, though, of course, with a very different ratio of GDP per capita.

Figure 20. Trend in Chinese GDP and fluctuation of inter-annual rates of growth between 1950-2008

Figure 20. Trend in Chinese GDP and fluctuation of inter-annual rates of growth between 1950-2008

Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows GDP values and the right axis, those of its inter-annual variation.

56What we mean to emphasize here is the link between urbanization activities which, in one stage constitute a substantial part of that annual flow (and therefore of employment and income) and in subsequent stages they do not. This may be illustrated more clearly by the proportion of gross fixed capital formation within the annual flow of wealth creation in Asia with respect to the world average (Figure 21).

57As shown there, then, towards 1975, the investment rate in Asia with respect to GDP reached the world average, which so far had outstripped the Asian rate. Since the mid nineteen seventies, such proportion has always been higher in Asia, though with marked discontinuities according to the urbanization and industrialization stages (cf. Asian trends in Figure 21). But as of the China effect, Asian investment has driven the proportion of gross fixed capital formation with respect to the product, also at the world level. This was clearly marked after 2001. Yet the global trend has kept decreasing over the last forty years. This could mean that, along each cycle of territorial expansion of the world industrial system, the driving forces of investment become progressively less dependent on urban expansion in purely quantitative terms, and tend to depend, instead, on the shift in the nature of existing activities, on their expansion or on those activities that are driven by the creation of new goods and services (technological innovation).

Figure 21. Trend in the quotient between gross fixed capital formation (buildings, and machinery and equipment) and annual GDP in Asia and at the world level. 1970-2011

Figure 21. Trend in the quotient between gross fixed capital formation (buildings, and machinery and equipment) and annual GDP in Asia and at the world level. 1970-2011

Note: Estimated on the basis of UNCTAD, UNCTADSTAT data.

58Figure 21 above shows that after 2002, the Gross Fixed Capital Formation/GDP quotient had a growing trend at the world level as a consequence of the China (or Asia in general) effect. This growth, however, was affected by the so-called financial crisis in 2009, which can be regarded as having a non-financial origin, as was suggested above. It can also be seen that investment on a global scale showed a decreasing and less productive trend between the end of the Golden Age and the 2001-2002 crisis. It grew slightly after the China effect, but it never reached its previous drive. This means that, in order to produce a new product unit, more investment was needed, or, in other words, capital was less productive.

Figure 22. Trend in world scale investment rate and in the incremental capital-output ratio. 1970-2011

Figure 22. Trend in world scale investment rate and in the incremental capital-output ratio. 1970-2011

Note: Estimated on the basis of UNCTADSTAT data. The quotient of GDP value at current and constant prices according to UNCTAD data series was used to estimate the 2005 value of Gross Fixed Capital Formation. The left axis shows the incremental capital/output ratio. The right axis shows the percentage of Gross Fixed Capital Formation over GDP on a world scale.

59So that what was stated above (that is, that capital was less productive) is not interpreted in terms of a decline in the rate of profit or a symptom of a terminal crisis of capitalism, Figure 23 shows the periodization of this phenomenon. Until approximately 1975, wealth creation went hand in hand with investment growth and the annual incremental output/annual incremental investment ratio was close to 4. After that came critical years when incremental investment and output fell or, even worse, output did not grow in spite of higher investment. It was the period of the implementation of Keynesian policies, which clearly showed that economic policy instruments were not enough to reactivate world economy (1979-1982). Between 1980 and 2002 this value reached a 6.5 average, that is, more or less the same incremental output with lower investment, which reveals weak wealth creation rates on a global scale and a high variability in the incremental capital-output ratio. It was the period when the North or the most developed countries grew at higher rates than the rest until they came to a significant stagnation (1998-2002). Then, when China became the factory of the world, the investment rate increased, and so did created global wealth, but to an average only three times higher than the new inflation, which clearly reveals the incidence of urbanization as a factor of wealth, as was the case in the Golden Age, though now attenuated by the total impact of developed economies.

Figure 23. Trend in investment rate at the world level and in the incremental capital-output ratio. 1970-2011, in millions of dollars of 2005

Figure 23. Trend in investment rate at the world level and in the incremental capital-output ratio. 1970-2011, in millions of dollars of 2005

Note: Estimated on the basis of UNCTADSTAD data. The quotient of GDP value at current and constant prices according to UNCTAD data series was used to estimate the 2005 value of Gross Fixed Capital Formation (GFCF). The left axis shows GDP and GFCF values, whereas the right axis shows the behavior of the incremental capital/output ratio.

60Unfortunately, there are not enough data to estimate the capital-output ratio on a world scale, but those available for developed nations show an interesting trend. In the first place, there was a marked decline between 1960 and 1973. Secondly, an increase was recorded between 1973 and 1983. Finally, there was a decrease between 1983 and 2001 (the OECD series was discontinued on that last year).

61An analysis of these data also reveals that 14 out of 22 countries show increases in their capital-output ratio at average values between 1976 and 2001 with respect to those recorded between 1960 and 1975, and that between 1960 and 1978, almost or over 60% of the capital was accounted for by the private residential sector and public investment, whereas towards 2000, this proportion was reduced to 38%. The residential investment sector, which accounted for between 26 and 27% of the total annual investment, reached only 13% in 2001, revealing a clear decreasing trend as of 1976 on. These data undoubtedly back up the arguments above.

62Figures 24 and 25 clearly show the structural differences in world economic growth between 1970 and 1979 with respect to 1980-2002 and 2003-2011. Lower Gross Fixed Capital Formation associated to lower urbanization rates will have serious effects on GDP composition, on investment conditions and recovery, and above all, on employment and the inclusion of more individuals in modern life.

Figure 24. Trend in annual average Gross Fixed Capital Formation and incremental world GDP, and the quotient between them by sub-periods. In constant dollars of 2005

Figure 24. Trend in annual average Gross Fixed Capital Formation and incremental world GDP, and the quotient between them by sub-periods. In constant dollars of 2005

Note: Estimated on the basis of UNCTADSTAT data. The quotient of GDP value at current and constant prices was used to estimate the 2005 value of Gross Fixed Capital Formation.

Figure 25. Trend in the average incremental capital/output ratio and in the standard deviation of the average by sub-periods (in constant dollars of 2005)

Figure 25. Trend in the average incremental capital/output ratio and in the standard deviation of the average by sub-periods (in constant dollars of 2005)

Note: Estimated on the basis of UNCTADSTAT data. The quotient of GDP value at current and constant prices was used to estimate the 2005 value of Gross Fixed Capital Formation.

63Then, though between 1980 and 2002 the world economy created more output per new unit of capital, this wealth creation was low on a global scale and very variable, unlike what happened when China became the factory of the world, in 2003-2012. Since this process is repeated, it cannot be understood without considering the impact of urbanization embodied in GDP.

64Thus, a description of the urban population growth process by large regions, expressed as nominal increases in urban population recorded over a five-year period between 1950 and 2010, reveals the following data. The rest of Asia reached its peaks towards 1985-1990, India continues close to a trend line, but China grew notably between 1980 and 2000, with impacts still over the trend in the last decade (Figure 26).

Figure 26. Increases in urban population by five-year periods between 1950 and 2010 by large regions. In thousands of individuals

Figure 26. Increases in urban population by five-year periods between 1950 and 2010 by large regions. In thousands of individuals

Note: Estimated on the basis of United Nations data (2012).

65Summing up, then, all the data shown here reveal mainly the measurable impact that the effect of China as the factory of the world has had along the first part of the twenty-first century regarding the following aspects: the creation of new wealth on a global scale, the proportion of industrial production in developing countries over the total industrial output and the reduction of the gap in average wealth per capita between developed countries and the rest of the world, and industrial activity in developed countries. However, what we want to emphasize here is that the strongest impact of this phenomenon derives from urbanization processes that, in turn, promoted industrial activity on a global scale. And that this was the consequence of two main factors: the demand for machinery and equipment related to the building of cities and infrastructure, and the impacts on the demand for raw materials, food and energy. All this activity and its multiplying mechanisms – for the terms of exchange were thus improved – has given rise to local industrial demand in many developing countries.

66Another issue has to do with the possibilities of the world economic system to create increasing levels of wealth. Evidence in this respect shows that this can be a very arduous process in the absence of extensive phenomena such as the incorporation of new urban population – as was the case in the last and a half decade. This was at least partially illustrated by what happened between 1980 and 2000. Over that period there were serious difficulties to create a level of annual wealth per capita enough for those cast out of the productive system to be included, a situation that keeps getting worse, whereas those who do not belong in this group enjoy a standard of living that no one would want to lose. This statement, which many would consider capitalist-oriented, may actually be typical of a system of wealth creation in which most of the annual flow of wealth is materialized as a non-reproducible stock of capital (the cities in themselves), though it is the condition for the extended reproduction of other activities.

  • 9 It was shown in Kozulj (2003) that the series of urban population growth out of phase with respect (...)

67Now if the hypothesis explained here is statistically verifiable, one would be tempted to say that there should be robust correlations between the increase in urban population and GDP growth. In this case, the first variable would be explanatory of the second, though both slightly out of phase with each other over time9.

68Figure 27 superimposes the series with values of urban population increases in large cities between 1975 and 1990 and the increase in GDP recorded between 1990 and 2008.

Figure 27. Increase in urban population (only in large cities) between 1975 and 1990 and increase in GDP between 1990 and 2008

Figure 27. Increase in urban population (only in large cities) between 1975 and 1990 and increase in GDP between 1990 and 2008

Note: Estimated on the basis of United Nations data (2012) and Maddison Historical GDP data. The right axis illustrates the increase in population in cities with over 750 thousand inhabitants between 1975 and 1990 expressed in thousands of individuals. The left axis shows the increase in GDP by countries where the cities with population increase between 1975 and 1990 belong, expressed in millions of dollars of 1990 for the period 1990-2008.

  • 10 Cf. Table 1, Annex 1 and the test carried out for the sake of this study.
  • 11 The White Test establishes if the explanatory variables of the model, their squares and all their p (...)

69A correlation can then be calculated between both variables: GDP increase as the dependent variable, and population increase in cities as the independent one. This reveals a high correlation coefficient (R2 = 0.70)10, which in itself does not add any information because these kinds of data show a high dispersion of values, and their results are, therefore, generally affected by heteroskedasticity (cf. Tables 1 to 4 and 19 to 23, Annex I). However, in correcting the variables in absolute value by dividing them by the square root of the explanatory variable in order to reduce the dispersion of the variance, a correlation of 0.60 results, which, in principle, satisfactorily exceeds the White test11, generally used to prove the existence of heteroskedasticity (cf. Tables 21 and 22, Annex 1.1).

  • 12 Cf. Tables 19 and 20, Annex 1, showing the result of the inverse hypothesis, that is, growth of pop (...)

70Another way of dealing with the problem of heteroskedasticity is to transform the absolute values of their variables into their natural logarithmic expressions (in this case, GDP increase and increase in population living in large cities in prior decades). In this case, the correlation decreases by 0.43 and results of the White test and of the Breusch-Pagan test show that in the logarithmic model and in that of the corrected variables, heteroskedasticity is solved. In all cases, a binary variable is added to differentiate developed countries from the rest, which, then, renders the explanatory variables statistically robust12.

71This, of course, does not mean to state that such simple models could explain the complex issues relating to development, or could be considered empirical evidence of the main hypothesis in this book. But I do consider that the results are interesting in themselves as regards the links between urbanization and growth from a point of view that is different from what the literature examined has considered.

72Back to the analysis of the data in Figure 27, then, several points of interest may be remarked, such as the cases of China and the United States, in the first place. But there are many other cases where two opposite situations are recorded. On the one hand, in developing countries with low average productivities, urban increase is much higher than the subsequent GDP proportional increase. On the other, there are paradigmatic cases where GDP increase is lower than the growth of large cities. This latter case is illustrated by countries such as Germany, Sweden, Norway, France and most of developed Europe, a leader in technological innovations which in turn are produced and consumed in many other cities, in their own markets and on a world scale.

73The residuals of the model are shown in Figure 1, Annex 1, where the weaknesses of such a simple explanatory model become evident, but this does not detract from its relevance in conceptual terms. That is, it shows that the construction of cities is a process embodied in GDP, whose relation with the creation of wealth as flow and as stock is particularly significant for the continuity of processes of extended reproduction of wealth. And it also shows that it is related to the causes of structural unemployment, destruction of human capital and the need to formulate specifically targeted policies, very different from the mere emphasis on macroeconomic policy instruments or on technological innovation.

74The correlation model improves significantly if other special cases are explicitly included in order to characterize the econometric model more accordingly to what we mean to express in theoretical and conceptual terms. These cases could be China, the United States, large developing countries with high levels of population growth in cities (such as Mexico, Brazil, India, among others), and developed countries. When this is done, the degree of correlation is R2 = 0.98 with acceptable statistical tests from the point of view of the relevance of each variable, though the efficiency of the model might be highly questionable on account of what has already been said about the dispersion of the absolute values of the variables (cf. Table 9, Annex 1). Of course, these results, whether corrected or in their logarithmic expressions, again produce lower values for the regression, and point out the methodological difficulties to contrast the hypotheses in a conclusive way.

75In turn, when the correlation involves urban population growth data of the period 1980-2000 to explain GDP growth between 1990 and 2008 (that is, with partially overlapping years in both variables, as is the case here with 1990-2000) and the series is also ordered, then the result is what is shown in Figure 28 below. This depicts the phenomenon more clearly.

Figure 28. Increase in urban population (only in large cities) between 1975 and 1990 and between 1980 and 2000, and increase in GDP between 1990 and 2008

Figure 28. Increase in urban population (only in large cities) between 1975 and 1990 and between 1980 and 2000, and increase in GDP between 1990 and 2008

Note: Estimated on the basis of United Nations data (2012) and of Maddison Historical GDP data. Right axis: increase in population in cities with over 750 thousand inhabitants in 2010 expressed in thousands of individuals (United Nations data). Left axis: increase in GDP by countries where the cities with population increase between 1975 and 1990, and between 1980 and 2000 belong, expressed in millions of dollars of 1990 for the period 1990-2008.

  • 13 In fact, in Annex 1 we show the existence of heteroskedasticity by means of the White test. As is w (...)

76In this case, the correlation increases with respect to the study with totally out-of-phase population (Table 1, Annex 1), with a resulting coefficient R2 = 0.88 (cf. Table 5, Annex 1). China would then disappear as a peculiar case in the analysis of residuals, which means that the model would adjust with no significant differences with respect to the only explanatory variable. Yet the efficiency of the model is low on account of what has been said about the dispersion of the variables, and the results to prove our hypothesis reveal the difficulties that have already been analyzed13.

77The above should not be taken literally, for China has also grown as a consequence of both the flow of foreign investment and its exports. However, it is precisely this argument that we wish to stress here: that it is not so important to determine which the initial trigger of growth was; the point is that it is embodied in urbanization and activities related to it. Once the urbanization process decelerates – which is expressed as the increase in absolute values of its population growth – many economic activities will likely enter a structural overcapacity stage some time later.

78Then reconverting industrial processes so that there are new products or new things to do is no simple task unless a kind of “sustainable re-urbanization” is promoted and there is a willingness to use the financial surplus generated by other activities – not so dependent on this cycle of long-term investment – to develop this process. This sustainable re-urbanization is to be promoted if the undesirable effects of creative destruction are to be minimized, for those effects would imply dismissing important installed capacities in the field of industries, human capital and technology.

79As will be seen in the second part of this book, an important number of activities could be promoted in relation to the prevention and mitigation of climate change – regardless of the causes of such change. These activities could provide employment and jobs compatible with this proposal. But this requires an understanding that sustainable development implies the strengthening of the right of present and future generations to enjoy a well-preserved environment with available resources, as well as their right to have jobs permitting them access to basic goods and services and to those deriving from present-day technological advance which result in higher levels of comfort.

80Without such options, the new things to do should be of equal or greater value in order to sustain growth. If not, the cycle is recessive and could be so for a long time once the limit of urban population increase has been reached on a global scale. This question is generally regarded in the context of the implementation of neo-Keynesian policies and their long-term efficiency, particularly after the 2009 crisis (Adam & Vines, 2009, pp. 532-535). Yet it is seldom seen from the physical or real point of view of the economy as conditioned by the structure of total installed capacity (that is, of the installed capacity of the supply of things to do and those that are already being done).

81This line of argument may be illustrated by correlating large groups of basic products with GDP. In that case, as has been stated, the resulting correlations are very high, though all of them affected by high levels of autocorrelation. Yet this is to be expected for each of these large products (or, as in the case of agriculture, a group of them) is contained in GDP measurement in a significant way for it is associated to products and activities which are in turn part of this wealth measurement.

Table 2. Value of the correlation between production of basic goods and GDP, indicating autocorrelation

Product R2 (explained by GDP) Value of the Durbin-Watson coefficient
Steel 0.92 0.15
Cement 0.95 0.11
Copper 0.99 0.28
Aluminum 0.98 0.24
Cereals 0.93 0.24
Agriculture 0.99 0.30

Note: Estimated on the basis of Maddison Historical GDP Data, Minerals Statistics and Information from the USGS, and FAO, Production Indices Cereals (Total). The Table shows urbanization stages and consumption of raw materials: autocorrelation in variables associated to GDP. Cases: energy, cement, steel, copper and food, and the results and residuals of the regressions. (Cf. Annex 1.2)

82The residuals of these correlations, though, reveal cycles associated with relations between urbanization and GDP growth, as argued here. Thus when comparing, for instance, agriculture and cement – both explained by GDP – the residuals show, in the case of cement, a very marked fall in the prediction with respect to GDP between 1965 and 1983, and an important recovery after the China effect. That is, this case goes alongside the incremental growth of urban population in the composition of wealth measurement. On the contrary, the production of cereals (and agricultural commodities in general) shows a larger share in relative terms after the waves of increase in urban population, and remains low in the explanation of new wealth when new waves of urbanization impact on the composition of growth. It is at least a possible interpretation along the lines of the arguments expressed here.

Figure 29. Behavior of the residuals of the correlation analysis of some basic products explained by GDP growth

Figure 29. Behavior of the residuals of the correlation analysis of some basic products explained by GDP growth

Note: Estimated on the basis of Maddison Historical GDP Data, Minerals Statistics and Information from the USGS, and FAO, Production Indices Cereals (Total). The left axis shows values for cement and cereals; the right axis, values for agriculture.

83What this actually shows is what was already presented in Figures 2 and 3 (Chapter 2) regarding how different stages of the urbanization process entail different growth dynamics in the demand for certain products (cf. Figure 29). Thus, if the composition of the creation of wealth as annual flow is conditioned by the magnitude of new urbanization processes and these get saturated, the role of the creation of new products to sustain the flow of wealth is complicated, irrespective of any improvement in business cycles, monetary and fiscal policies meant to affect aggregate demand.

84The results of several of these correlations between behavior of products and GDP are presented in Annex 1. Now this approach to the question has serious implications on employment conditions and, therefore, on the insertion of individuals in society, fully entitled to access to goods and services.

85To close this chapter, rich in quantitative data, it is necessary to point out that there are other ways of measuring the links between urbanization and growth, such as urbanization percentage vs. GPD per capita. Though properly reflecting the interrelation between both, such measurements do not take into account what happens with the destruction of capital inherent in the urbanization process as included or embodied in GDP measurement. In spite of that, the results deserve attention.

86In this respect, Table 3 shows the values of the correlation between GDP per capita and percentage of urban population with data for 47 countries, expressed as natural logarithms of the absolute values of those data.

Table 3. Value of the correlation between the logarithm of GDP per capita for each country and the logarithm of the urbanization percentage reached: synchronic data 1950-2010

Year (Synchronic with Log GDP/capita Value of R2 Value of the coefficient Log urban population percentage Value of Student’s t-test Autocorrelation (Durbin Watson)
1950 0.605 0.729 13.90 1.58
1970 0.630 1.014 14.70 1.4
1990 0.537 1.295 12.95 1.08
2010 0.405 1.530 9.94 1.2

Note: Estimated on the basis of Maddison Historical GDP Data and United Nations data (2012). In these cases, the problem of heteroskedasticity found in previous exercises is solved, but the precedence of rural-urban migration processes as an explanatory factor of the subsequent growth of the total output is not clearly reflected. Cf. Annex 1, Tables 13 to 18.

87These econometric studies including only two variables reveal an increase in the value of the correlation between 1950 and 1970 (from 0.60 to 0.63), and a subsequent decrease towards 1990 and even more towards 2010 (from 0.54 to 0.40). Likewise, it is possible to observe a decrease in the elasticity of GDP per capita with respect to the urbanization level, a relative increase in the autocorrelation and a decrease in the significance of the coefficient, though always showing highly significant values.

88This analysis of the links between urbanization and growth also reveals interesting data, particularly when considering the residuals and their variation for the specific cutoff years, such as 1950, 1970 and 2010, as shown in Figures 30 and 31 below.

Figure 30. Trend in the residuals of synchronic correlations between urban population percentages and of GDP per capita. Contrasted data for 1950 and 2010

Figure 30. Trend in the residuals of synchronic correlations between urban population percentages and of GDP per capita. Contrasted data for 1950 and 2010

Note: Estimated on the basis of Maddison Historical GDP Data and United Nations data (2012). Cf. Annex 1, Tables 13 to 18.

  • 14 It should be remembered, for example, that in 1793 the East India Company traded only with Guangzho (...)

89The first Figure reveals that departures from the correlation model in 1950 are smaller with respect to those recorded in 2010, and with clearly identified cases both above and below the model. For example, all oil exporting countries or those with highly valuable natural resources on the international market already appeared above the correlation model between GDP per capita and urbanization percentage in 1950, and so did developed countries in Europe, North America and some cases such as Australia and New Zealand. Likewise, the large Asian countries showed departures below the model in the same correlation. It should be remembered that in 1950, most of these countries were emerging from colonial systems or from strong dependence on Great Britain (India, China14, Pakistan, Singapore, just to mention the paradigmatic cases).

90It is curious to observe that very large countries reveal intermediate or very low departures from the average when ranked from lowest to highest according to the level of the residuals with data for 2010. That is, their increase in GDP per capita has been alongside their urbanization level – a very different way of considering the cases of China and India and their own internal impact of the urbanization process expressed by means of out-of-phase increases in urban population with respect to GDP.

Figure 31. Trend in the difference between residuals of synchronic correlations between the values of natural logarithms of urban population percentage and of GDP per capita. 1950, 1970 and 2010

Figure 31. Trend in the difference between residuals of synchronic correlations between the values of natural logarithms of urban population percentage and of GDP per capita. 1950, 1970 and 2010

Note: Estimated on the basis of Maddison Historical GDP Data and United Nations data (2012). Cf. Annex 1, Tables 13 to 18.

91A new light is cast on this analysis when the trend in the difference in value of the residuals of the correlations between urbanization rate and GDP per capita is added (Figure 31). In the first place, this difference was much smaller between 1950 and 1970 (values of its natural logarithms) than between 1950 and 2010. This means partly that the correlation was stronger – and so is clear from the results – between 1950 and 1970 than after that period (Table 3).

92At the same time, this is coherent with the argument that economic growth in developing and developed countries was boosted by the wave of post-war urbanization and that after that (the period of the Glorious Thirty, the Golden age, the end of Fordism), the driver of growth focused rather on technological innovation and the creation of new things to do, even when those already done became obsolete as a result – a clear example of creative destruction. This period also coincided with the strong concentration of industrial production in developed countries, which went hand in hand with a partial de-industrialization of developing ones. In fact, this is what has been shown with irrefutable data by comparing the contribution of developing and developed countries to industrial growth, and more precisely, between 1980 and 2000/2002. Furthermore, the main emphasis has been on the fact that the most disruptive element towards the end of those years was the drive of the new wave of urbanization, basically that in Asia (with China becoming the factory of the world).

93These facts are therefore central because the dynamics of the economy, the growing creation of new flows of wealth, is linked to the creation and expansion of urban centers. This promotes many industrial and service activities that are embodied in the measurement of that flow of wealth (GDP), activities which in turn are materialized as stock of wealth. A part of this, though, does not generate those activities, at least not in equivalent magnitudes. As has been said, this is a crucial issue regarding the possible destruction of human capital and also regarding the need to find new things to do with values equal or higher than those of the things no longer done.

94Can the global market – or the local market within each country – ensure this? Could the State do it? Reasonable doubt has already been expressed in this respect so that such options should not be considered possible, for they are polarized, too ideological, and not scientific enough. And this is so because the origin of the problem has neither been understood nor sufficiently explained. That this is not simply the opinion of the author is illustrated by the following quote and subsequent digression:

“Globalization, a dominant force in the 20th century’s last decade, is shaping a new era of economic growth of nations. It was anticipated that with high rates of economic growth, the incidence of poverty would reduce. But this has not happened. The global economy has also fragmented production processes, labour markets, political entities and societies. So, while globalization has positive, innovative, dynamic aspects - it also has negative, disruptive and marginalizing aspects.

A critical feature of globalization is new lines and forms of stratification between places, people and groups. In particular, it is manifested in much greater income inequalities. In all the regions, where the absolute number of poor has increased, a majority of them are in urban areas that have been the key drivers of the global economy. […] the urban poor are the worst affected group when there is sudden decline in economic growth. The urban poor, unlike the rural poor, are the most vulnerable group because most national governments in developing countries do not provide any social safety nets for them”. (Metha, 2000)

95Since it is Pratibha Mehta, the Coordinator of the United Nations Human Settlements Program, who actually admits this, it is curious that he cannot think of concrete answers to this social fragmentation and vulnerability of the urban poor in a context of economic decline, as could be programs for them to focus on the global construction of a sustainable habitat. This could channel the surplus of savings over investment possibilities, and also narrow the gap now existing between labor supply and demand.

  • 15 Cf. the research carried out by Peter Diamond, Dale Mortensen and Christopher Pissarides, winners o (...)

96It is well known that one of the main problems of the labor market in all countries, but even more so in developing ones, is that labor force supply and demand do not behave the way the classical theory supposes, that is, as buyers and sellers instantly brought together, at no cost, with perfect information about all the prices of goods and services. But the fact that such statement was made by the Economic Sciences Nobel Prize Committee in 2010 about those who discovered that unemployment subsidy-oriented policies simply prolong unemployment shows at least how rudimentary the Economic Science still is when implementing policy instruments. And it also shows the futility of the debate between free market and statism, which the press echoes without going deep into any of the structural causes of poverty and unemployment15. Another quote may better illustrate this point:

  • 16 Our translation from the author’s quote of Galindo Martín (2010).

“The crisis affecting the economy over the last years is having a very negative effect on unemployment. The Governments have implemented measures to cope with this problem, and the results have not been as positive as expected. On the basis of the idea that economic growth brings about employment, policies were promoted to boost it which, indirectly, affect employment. Another option has been to act directly on the labor market to improve the situation. In this respect, the old controversy between neo-classical and Keynesians has revived. The former accuse the latter of being responsible for unemployment, as a consequence of the expansive fiscal measures and regulations that have introduced unwanted rigidities, so that they demand more flexibility for the labor market. The Keynesians, on the other hand, consider that the problem is the insufficient aggregate demand and recommend implementing expansive policies. The Swedish Academy, in turn, granted their award to three economists who analyzed the problems of the labor market and the existence of rigidities in it. The origins of those rigidities may be of a very diverse nature, such as incomplete data, cost of transport, among others, so that they question some of the contributions of the neo-classical theory, though they also criticize, as will be seen below, some Keynesian recommendations […] The conclusions drawn out of the DMP (Diamond-Mortensen-Pissarides) model regarding employment policies have to do with the fact that an increase in wages supposes a reduction in the creation of jobs, more unemployment and higher real wages, as well as an increase in the interest rate, only that instead of increasing, it reduces the real wage. In turn, the introduction of technology reduces unemployment, as does an increase in productivity. Finally, it is stated that the more benefits and protections are introduced in the market, the longer unemployment will prevail”. (Galindo Martín, 201016)

97It is not difficult to see that, whether as an interpretation of the DMP model, or as an economic policy recipe – and without considering the structural causes leading to both lack of job vacancies and growing unemployment – the above quote may not be related to wages or to monetary and fiscal policy measures, but to a serious mismatch between the abilities of the unemployed and those required by employers. These, no doubt, have to do with what is done, and with what things those who no longer find a market for what they can do well can really do. Is it perhaps logical that an evolutionary explanation such as is suggested here be disregarded in a model trying to explain this reality? My answer is that it is not, and even less so in view of the vast volume of information available.

98But even more serious is that the language used to explain the problem is, to say the least, somewhat naïve. Consider, for instance:

  • 17 The translation is ours.

“Why are there so many unemployed people, while there is an important number of job offers? How can the economic policy affect unemployment? These are questions that those awarded with the Nobel Prize in Economic Sciences 2010 try to answer. Though their theory may apply to different markets, it is the labor market that they have focused on. The classical theory of the market says that buyers and sellers are instantly brought together, at no cost, with perfect information about all the prices of goods and services and prices are determined so that supply equals demand, there are no surpluses of supply or demand and all the resources are fully used. But this is not what happens in real life. High costs are often the consequence of buyers not finding sellers, and vice versa. Even after these two actors are brought together, the products in question may not correspond with the buyers’ demands. A buyer could consider the seller’s price too high, or a seller could consider a buyer’s supply too low. Then, there will be no transaction”. (Carbajal Suárez & Alonte, 201017).

99In the case of the labor market, the phrase: “Even after they are brought together, the products in question may not correspond with the buyers’ demands” will be referring to the people finding a job and the business people and providers of jobs, respectively. It is clear that it cannot mean only the wage question. Labor and labor demand are obviously not homogeneous variables. The question of human capital formation is implicit here. For example, rural-urban migrants who no longer get jobs in activities related to the construction, have abilities that are not likely to be sold in this sector and, if they do not quickly acquire new abilities, their situation in the market gets very precarious. But let us suppose that they can really do some public work (e.g., as flagmen in road construction), or can work as garden, park, building, house maintenance people, or as peddlers. In that case, if it is up to the market to decide, it is likely that supply will outstrip demand. If the expenses incurred are afforded by the State via public spending, the incidence of this on expenditure and the necessary fiscal pressure will not be minor issues in macroeconomic policy, but how willing are the citizens to pay for responsible social inclusion? Unemployment insurance may cover expenses of the urban poor in need of it, and at the same time perpetuate unemployment and maintain a certain level of aggregate demand. But it may also get people (and whole generations) used to the idea that having access to goods and services is a universal right even when one does not contribute anything to the productive process for, ultimately, it is the system that does not generate employment opportunities for them to contribute to that process.

  • 18 John F. Nash (1928-2015) was an American economist and mathematician. He is well known for his extr (...)

100Critiques of the DMP model by authors such as Robert Shimer (2005), for instance, formally question the labor market equilibrium model (determining the wage level), its theoretical vs. empirical bases for a certain type of recessive and very pronounced cycle and also the Nash18 bargaining assumptions as probably inadequate. In my opinion, such critique is still trapped in formalisms deriving from supposing that the problem lies in the inflexibility of the labor market, for at no point do they deal with highly segmented labor markets, and thus the model cannot correctly explain concrete facts, the empirical side of the question. Though without explicitly meaning it, maybe, such critique reinforces the idea that the rigidity of the labor market is the ultimate cause of unemployment, which is known to be stated by the neoclassical, who always suppose there are an infinite number of things to do which must or may have a market.

101If this is so, the problem ends up being that “those poor people are not creative enough, are mentally or physically lazy, or lack enterprising drive”. Yet, even when some of those epithets on the real poor might be partially true – or at least applicable to a subgroup of them –, what these distinguished academics seem to ignore is that these people migrated from the country to the cities when there were many things to do in them, for the cities were being built at a very fast pace, and in those stages certain specific markets required labor force. The wage level in that context – of course determined by conditions both objective and subjective – was surely higher than they could earn in primary activities in the rural areas where they came from.

102What is not said in these analyses is that, at a certain point, this labor force is no longer needed in the cities, but then the decisions the people have made admit no turning back to previous conditions, whether because of objective causes (loss of the land they owned, family disintegration), or subjective ones (once they have had a taste of life in the city, country life might seem unappealing, and it will be worth struggling to stay there, whatever the cost). This may have irreversible negative impacts on future generations in some countries regarding the loss of human capital, which is only seen in the long term. Thus the static model proves unhelpful.

103Then, if the link between destruction of human capital and the urbanization process is not stressed as the original and inevitable cause of an evolutionary process, the need to anticipate the implementation of labor reconversion programs will not be considered. Nor will a program of public and private investments agreed upon by the State and private actors be seen as a necessary part of this process.

104Thus everybody somehow ignores – or takes it too much for granted to make it explicit – that falls in investment may be caused by null or negative cost-effectiveness expectations due to the fact that important markets get saturated and there are no others that could attractively replace them. When such potential markets do turn up, they require abilities that are scarce in a segmented labor market and are difficult to attain in the short term without educational institutions up to the challenges of the twenty-first century.

105The market solution relies too much on the idea of innovation as a panacea rather than as a trap. Most public policies stress innovation, the promotion of people with an enterprising drive, and the like. Very few consider adapting the supply of public and private services to easier – albeit less cost-effective – tasks that will enable people to make a living and societies to be less “marginalizing”. This will be dealt with below. Suffice it to say here that when there is no knowing what the product is made of in each stage of the urbanization process and what other new things to do are compatible with what people can do or can learn to do in the short term, the battle for a more marked growth as a way of reducing poverty is lost beforehand. And so is the utopia of stopping any form of growth without provoking an increase in extreme poverty and the destruction of what humankind has achieved over the last two or more centuries – which the promoters of zero growth seem to ignore.

106Back to Figures 30 and 31, another peculiarity has to do with which countries have grown above what their urbanization rates would have predicted on the basis of a linear correlation and which have grown below such “prediction”. On the upper right corner of Figure 31 are countries whose economic growth, as time goes by, can no longer be explained on the basis of their incremental urbanization rate (the United States, Switzerland, Canada, Japan, Norway, Denmark, Sweden, France, Hong Kong, The Netherlands, Belgium, Australia, Austria, Finland, the United Kingdom, Italy, Germany, in that order), but also those small countries benefited by exceptional oil market conditions (Trinidad and Tobago, Equatorial Guinea). What they have in common is their insertion as leaders in technological innovation – which has also been described as a characteristic of the hundred leading cities in competitiveness and innovation – and the fact that they are oil exporters.

107On the lower right corner, on the other hand, are the less successful cases within the scarcely urbanized countries. Many of them are in Africa, and all of them reveal values of the residuals that are negative or inverse of those of developed countries. This means that their GDP per capita was smaller than the likely average in the correlation with the urban population percentage both between 1950 and 1970, and between 1950 and 2010, but even more so between 1970 and 2010. Yet, apart from these exceptions, there are many countries whose growth is explained by their level of urbanization, even if this has been the result, in turn, of a competitive advantage based on an income-producing natural resource at some point in history.

108Summing up, then, this calls for a consideration of the drivers of growth as different according to periods and regions, but above all, of the role of technological innovation, its limits, what new things to do are necessary even when the market may not be willing to pay for them, and which will turn up out of the market itself. Of course such a solution requires State intervention in each nation, but particularly an international agenda with clear aims as to how to implement orderly labor, training, financial, technological, cultural and urban habitat policies. It is then clear that this is not compatible with capitalism as it is now known, and neither is it to be solved by socialist policies past and present unless consumer ambitions are reduced by means of the individuals’ decision or by the abolition of certain human rights.

109Before going into the proposals – to some extent a proactive issue – another aspect needs to be explained. It has to do with why technological innovation, though an essential part of the transformation process, may also be a trap preventing the achievement of more sustainable development on a global scale.

Notes

1 Or 2013/2014, when data are available.

2 Basically, all those institutions belonging to the UN system, such as the IMF, the WB, regional bodies, UNIDO, UNDP. Also national development banks, national state agencies aiming to promote industrialization, among others.

3 Cf. Zizek, S. (2002, p. 22).

4 Austria, Belgium, Denmark, Finland, France, Germany, Italy, The Netherlands, Norway, Sweden, Switzerland and the United Kingdom.

5 When this part of the book was written, certain crises could not yet be foreseen in all their crudeness. This is the case of the crisis that affected the BRIC bloc (made up of Brazil, Russia, India, China and South Africa), recorded between 2015 and 2016, which is not only economic but a real turn in political leadership, particularly in Russia and Brazil. Another example is the stock-exchange crisis in China.

6 UNCTAD Stat, “GDP by type of expenditure and Value Added by kind of economic activity, annual, 1970-2011, US Dollars at current prices and current exchange rates in millions”.

7 This explanation is neither untimely nor ad hoc, as a review of the article in question will reveal (Kozulj, 2003), with particular attention to Figure 4.

8 Heteroskedasticity is usually produced when working with cross-sectional series containing samples of very different sizes. Its consequences may invalidate the results of the regressions in several different ways. However, our statement is based on the precedence of a variable with respect to another one in synchronic tests, which is why the best results in terms of the regression coefficient are all affected by the same dispersion of the values of the variables in question. It is, then, only a test, not conclusive evidence of the hypothesis put forward in this book. However, the test with variables corrected with approximate solutions to achieve homoskedasticity reveals acceptable results in the same logical direction, that is, that urban population increase precedes GDP growth.

9 It was shown in Kozulj (2003) that the series of urban population growth out of phase with respect to increases in the product some decades later behave better than if the explanatory variable of urban population growth or growth in large cities was “explained” by the prior, subsequent or synchronic growth of the product.

10 Cf. Table 1, Annex 1 and the test carried out for the sake of this study.

11 The White Test establishes if the explanatory variables of the model, their squares and all their possible unrepeated crosses are useful to determine the trend of the square of the error. That is, if the trend of the explanatory variables and their variances and covariances are significant to determine the value of the sample variance of the errors, understood as an estimate of the variances of the random disturbances. In principle, the R2 as a proportion of the variance of the real endogenous variable explained by the estimated variable should be very small if the explanatory capacity of the regressors under consideration is also very small; because of their construction, such regressors would be representative of the variances and covariances of all the explanatory variables of the original model. Then, a value of R2 that is small enough will evidently lead to the conclusion that there is no heteroskedasticity in the model produced by the values of the explanatory variables considered in the initial model. In this case, using the White test for the corrected values results in R2 = 0.18, which indicates that the problem posed by the large dispersion of the data in the original model has been satisfactorily solved.

12 Cf. Tables 19 and 20, Annex 1, showing the result of the inverse hypothesis, that is, growth of population in citites explained by the subsequent generation of wealth.

13 In fact, in Annex 1 we show the existence of heteroskedasticity by means of the White test. As is well known, one of the simplest forms of approaching this problem is to transform the data into the logarithmic functional form. In this case, the results show a smaller correlation, but estimators appear unbiased and consistent, and exceed such test. Interested readers might want to refer to Gujarati & Porter (2009), and Carrascal (2001, p. 227).

14 It should be remembered, for example, that in 1793 the East India Company traded only with Guangzhou, but the Opium War soon started…, which made it influential all around China. Cf. http://www.economist.com/news/essays/21609649-china-becomes-again-worlds-largest-economy-it-wants-respect-it-enjoyed-centuries-past-it-does-not

15 Cf. the research carried out by Peter Diamond, Dale Mortensen and Christopher Pissarides, winners of the Nobel Prize in Economic Sciences 2010, in Albrecht, (2011)

16 Our translation from the author’s quote of Galindo Martín (2010).

17 The translation is ours.

18 John F. Nash (1928-2015) was an American economist and mathematician. He is well known for his extraordinary talent for mathematical analysis. He researched into the game theory, and was awarded the Nobel Memorial Prize in Economic Sciences in 1994 on account of that, which he shared with John Harsanyi and Reinhard Selten. He was the first to define bargaining games, around 1950, as a set of likely assignments of utilities resulting from all the likely agreements that the bargaining parties may reach, and an assignment corresponding to the gains that each of the parties gets in case an agreement cannot be reached. In order to find a solution to the bargaining problem, he resorts to postulating a series of desirable properties (axioms) that such a solution should satisfy, and he then goes on to define it. In this context, a bargaining solution is a rule meant to share the gains of any bargaining problem. Nash, then, introduces four axioms: great efficiency (Pareto optimality); symmetry; invariance to equivalent utility representations and independence of irrelevant alternatives. And he shows that the solution he offers – that is, the one which maximizes the product of the parties’ surplus utilities – is the only one satisfying all four axioms.

Table des illustrations

Titre Figure 5. Growth of World Gross Product: developed and rest of the world between 1950 and 2008
Légende Note: Estimated on the basis of Maddison Historical GDP Data. Values expressed in “Geary-Khamis dollar” (GK$), a hypothetical unit of currency that has the same purchasing power parity that the U.S. dollar had at a given point in time – in this case, 1990. Comparison of this long series with UNCTAD 1970-2008 data behaves in a similar – though not identical – manner.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-1.png
Fichier image/png, 43k
Titre Figure 6a. Growth in Gross World Product per capita: developed countries and rest of the world between 1950 and 2008 (equivalent to dollars of 1990)
Légende Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows the trend in GDP per capita in GK$ per capita The right axis shows the value of the coefficient between such GDP per capita in the most developed nations with respect to the same average indicator in the rest of the world.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-2.png
Fichier image/png, 211k
Titre Figure 6b. Growth in Gross World Product per capita: developed countries and rest of the world between 1970 and 2013 (in constant dollars of 2005)
Légende Note: Estimated on the basis of UNCTAD data of GDP at constant 2005 values, updated 1970-2012 population series, and 2009-2013 rates according to United Nations projections. The left axis shows the trend in GDP per capita in dollars of 2005 per capita. The right axis shows the value of the coefficient between such GDP per capita in the most developed nations with respect to the same average indicator in the rest of the world.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-3.png
Fichier image/png, 205k
Titre Figure 7a. Growth of incremental Gross Domestic Product per capita: developed countries and rest of the world between 1950 and 2008 (values equivalent to dollars of 1990)
Légende Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows the annual variation of GDP per capita in GK$ per capita (lines). The right axis shows the value of the coefficient between such variation in annual GDP per capita in the most developed nations and the same average indicator in the rest of the world (bars).
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-4.png
Fichier image/png, 67k
Titre Figure 7b. Growth of Incremental Gross World Product per capita: developed countries and rest of the world between 1970 and 2013 (in current dollars)
Légende Note: Estimated on the basis of UNCTAD data, according to GDP at constant 2005 values and 1970-2009 updated population series, and 2009-2013 rates using United Nations projections. The left axis shows the annual variation of GDP per capita in dollars of 2005 per capita (bars and dark line). The right axis shows the value of the quotient between such annual incremental GDP per capita in the most developed nations and the same average indicator in the rest of the world (dashes).
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-5.png
Fichier image/png, 62k
Titre Figure 8. Trend in product per capita between 1950 and the 2009 crisis: values in the series with annual variations of GDP per capita in three long periods
Légende Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows the trend in GDP per capita in GK$ per capita. The right axis shows the value of the annual increase in GDP per capita expressed in the same unit.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-6.png
Fichier image/png, 129k
Titre Figure 9. Increase in wealth creation and in total and urban population and population in cities with over 750 thousand inhabitants. Values in thousands of individuals, in GK$ and in GK$ per every thousand individuals by five-year periods
Légende Note: Estimated on the basis of United Nations 2012 data, Maddison Historical GDP Data, UNCTAD, UNCTADSTAT, to connect GDP growth by 2012. The left axis shows the values of the five-year period increase in total, urban and large cities population in thousands of people (bars). The right axis shows the value of the quotients between the five-year period increase in GDP and the increase in population according to the same classification in 1990 GK$ values.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-7.png
Fichier image/png, 75k
Titre Figure 10. Population increase in cities with over 750 thousand inhabitants and in urban population with respect to total population increase
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-8.png
Fichier image/png, 197k
Titre Figure 11. Population increase in cities with over 750 thousand inhabitants and in urban population in continental China. Historical data for 1950-2010 and projections by 2025 in thousands of individuals
Légende Note: Estimated on the basis of United Nations 2012 data. The left axis shows population values by urban and large cities population (bars). The right axis shows the values of the increase in such population by five-year periods.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-9.png
Fichier image/png, 62k
Titre Figure 12. Trend in foreign investment in China in annual million dollars, as a percentage of GDP and as accumulated stock without applying depreciation
Légende Note: Estimated on the basis of UNCTAD 2014 data. The left axis shows the percentage accounted for by accumulated foreign investment received by China in relation to GDP. The right axis shows annual and accumulated investments without depreciation.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-10.png
Fichier image/png, 35k
Titre Figure 13. Trend in total exports by China and the United States of America. Products, and Goods and Services in millions of current dollars between 1980 and 2012
Légende Note: Estimated on the basis of UNCTAD data.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-11.png
Fichier image/png, 79k
Titre Figure 14. Trend in annual variations of China’s imports between 1995 and 2009 (value of the base 100 index in 2000)
Légende Note: Estimated on the basis of UNCTAD data.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-12.png
Fichier image/png, 55k
Titre Figure 15. Trend in annual variations in China’s imports of machinery and equipment between 1995 and 2009 and in Industry Value Added in developed countries (value of the base 100 index in 2000)
Légende Note: Estimated on the basis of UNCTAD data.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-13.png
Fichier image/png, 74k
Titre Figure 16. Trend in global growth rates of world economy between 1950 and 2012
Légende Note: Estimated on the basis of Maddison Historical GDP Data and UNCTAD, UNCTADSTAT, to connect GDP growth by 2012. The left axis shows world GDP in GK$. The right axis shows the percentage of the rate of variation in GDP year by year.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-14.png
Fichier image/png, 117k
Titre Figure 17. Trend in annual variations in Industry Value Added (IVA) in developed and developing countries and on a world scale (1970-2011). In millions of current US dollars
Légende Note: Estimated on the basis of UNCTADSTAT data. Chosen variable: annual changes in the absolute value of Industrial Value Added (IVA) at current prices and according to current exchange rates in the countries in question.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-15.png
Fichier image/png, 82k
Titre Figure 18. Trend in accumulated variations in value added by large sectors in developed and developing countries. Data shown as the difference between 2003-2011 in millions of current US dollars
Légende Note: Estimated on the basis of UNCTADSTAT data, 1970-2011.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-16.png
Fichier image/png, 31k
Titre Figure 19. Trend in GDP in the United States and in the fluctuation of inter-annual rates of growth in 1916-2008
Légende Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows GDP values and the right axis, those of its inter-annual variation.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-17.png
Fichier image/png, 247k
Titre Figure 20. Trend in Chinese GDP and fluctuation of inter-annual rates of growth between 1950-2008
Légende Note: Estimated on the basis of Maddison Historical GDP Data. The left axis shows GDP values and the right axis, those of its inter-annual variation.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-18.png
Fichier image/png, 678k
Titre Figure 21. Trend in the quotient between gross fixed capital formation (buildings, and machinery and equipment) and annual GDP in Asia and at the world level. 1970-2011
Légende Note: Estimated on the basis of UNCTAD, UNCTADSTAT data.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-19.png
Fichier image/png, 104k
Titre Figure 22. Trend in world scale investment rate and in the incremental capital-output ratio. 1970-2011
Légende Note: Estimated on the basis of UNCTADSTAT data. The quotient of GDP value at current and constant prices according to UNCTAD data series was used to estimate the 2005 value of Gross Fixed Capital Formation. The left axis shows the incremental capital/output ratio. The right axis shows the percentage of Gross Fixed Capital Formation over GDP on a world scale.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-20.png
Fichier image/png, 279k
Titre Figure 23. Trend in investment rate at the world level and in the incremental capital-output ratio. 1970-2011, in millions of dollars of 2005
Légende Note: Estimated on the basis of UNCTADSTAD data. The quotient of GDP value at current and constant prices according to UNCTAD data series was used to estimate the 2005 value of Gross Fixed Capital Formation (GFCF). The left axis shows GDP and GFCF values, whereas the right axis shows the behavior of the incremental capital/output ratio.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-21.png
Fichier image/png, 88k
Titre Figure 24. Trend in annual average Gross Fixed Capital Formation and incremental world GDP, and the quotient between them by sub-periods. In constant dollars of 2005
Légende Note: Estimated on the basis of UNCTADSTAT data. The quotient of GDP value at current and constant prices was used to estimate the 2005 value of Gross Fixed Capital Formation.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-22.png
Fichier image/png, 144k
Titre Figure 25. Trend in the average incremental capital/output ratio and in the standard deviation of the average by sub-periods (in constant dollars of 2005)
Légende Note: Estimated on the basis of UNCTADSTAT data. The quotient of GDP value at current and constant prices was used to estimate the 2005 value of Gross Fixed Capital Formation.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-23.png
Fichier image/png, 93k
Titre Figure 26. Increases in urban population by five-year periods between 1950 and 2010 by large regions. In thousands of individuals
Légende Note: Estimated on the basis of United Nations data (2012).
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-24.png
Fichier image/png, 58k
Titre Figure 27. Increase in urban population (only in large cities) between 1975 and 1990 and increase in GDP between 1990 and 2008
Légende Note: Estimated on the basis of United Nations data (2012) and Maddison Historical GDP data. The right axis illustrates the increase in population in cities with over 750 thousand inhabitants between 1975 and 1990 expressed in thousands of individuals. The left axis shows the increase in GDP by countries where the cities with population increase between 1975 and 1990 belong, expressed in millions of dollars of 1990 for the period 1990-2008.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-25.png
Fichier image/png, 338k
Titre Figure 28. Increase in urban population (only in large cities) between 1975 and 1990 and between 1980 and 2000, and increase in GDP between 1990 and 2008
Légende Note: Estimated on the basis of United Nations data (2012) and of Maddison Historical GDP data. Right axis: increase in population in cities with over 750 thousand inhabitants in 2010 expressed in thousands of individuals (United Nations data). Left axis: increase in GDP by countries where the cities with population increase between 1975 and 1990, and between 1980 and 2000 belong, expressed in millions of dollars of 1990 for the period 1990-2008.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-26.png
Fichier image/png, 73k
Titre Figure 29. Behavior of the residuals of the correlation analysis of some basic products explained by GDP growth
Légende Note: Estimated on the basis of Maddison Historical GDP Data, Minerals Statistics and Information from the USGS, and FAO, Production Indices Cereals (Total). The left axis shows values for cement and cereals; the right axis, values for agriculture.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-27.png
Fichier image/png, 92k
Titre Figure 30. Trend in the residuals of synchronic correlations between urban population percentages and of GDP per capita. Contrasted data for 1950 and 2010
Légende Note: Estimated on the basis of Maddison Historical GDP Data and United Nations data (2012). Cf. Annex 1, Tables 13 to 18.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-28.png
Fichier image/png, 96k
Titre Figure 31. Trend in the difference between residuals of synchronic correlations between the values of natural logarithms of urban population percentage and of GDP per capita. 1950, 1970 and 2010
Légende Note: Estimated on the basis of Maddison Historical GDP Data and United Nations data (2012). Cf. Annex 1, Tables 13 to 18.
URL http://books.openedition.org/eunrn/docannexe/image/3363/img-29.png
Fichier image/png, 78k

CC-BY-NC-ND-4.0

Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search