Cost effects of the latest mining boom on emerging economies

ABSTRACT An input–output approach is used to identify the impact of the last mining boom (2000–08) on the cost and prices of mining industry products in four BRICS countries (Brazil, Russia, India and China). Using input–output tables and environmental information from the World Input–Output Database (WIOD), the analysis points to higher costs in emerging economies due to higher wages, slower technical progress, positive pass-through of oil shocks and partial sequential depletion processes in industrial minerals.


INTRODUCTION
In recent times, the extraction levels and prices of mineral resources have increased to an extent rarely seen before. The rising price record has been one of the longest and broadest since the Second World War, which poses the question as to what factors explain this mining price boom. The aim of this study is to quantify the cost effects for emerging economies of the mining boom between the years 2000 and 2008 in some BRICS countries, specifically Brazil, Russia, India and China, using an input-output approach. Price indicators were built using a cost-push model (Miller & Blair, 2009), whereas productivity and depletion indicators focused on a subsystem approach (Pasinetti, 1973). Input-output data and environmental information were drawn from the World Input-Output Database (WIOD). The study provides strong evidence of higher costs resulting from the specific characteristics of emerging economies, such as higher wages, slower technical progress, pass-through of oil shocks and partial sequential depletion of industrial minerals.
The rest of this article is structured as follows. Section 2 reviews the literature on the recent mining price boom. Section 3 describes the methodology and indicators used. Section 4 presents and critically evaluates the empirical analysis results. Finally, Section 5 summarizes and concludes.

LITERATURE REVIEW
The turn of the century witnessed major transformations in the world of natural resources due, in particular, to China's high demand for fossil fuels and minerals and the resulting sharp increase in production and imports (Erten & Ocampo, 2013).
Other factors of the mining transformation were the acceleration of long-term changes. Particularly, the shift of mineral deposit locations from developed to less developed countries has been a trend since the beginning of the 20th century. However, in the last decades, production has been concentrated in countries of Latin South America, Africa and Asia (Gorenstein & Ortiz, 2018), with debatable effects in their economic performance (Scholvin et al., 2021). The main reasons for the new locations were a new combination of technical and physical characteristics of the active deposits (Ericsson & Hodge, 2012).
Almost all the existing literature (cf. Erten & Ocampo, 2013) has focused on the demand side and the critical role of financial investments (Gorenstein & Ortiz, 2018). Just a few papers deal with the reaction of the supply side (Radetzki et al., 2008), focusing only on productivity. On this topic and despite the slow response in greenfield exploration, the productivity of active mines has not played a dynamic role in recent times (Bartos, 2007). In particular, this scenario has been observed in the case of oil (Clifford et al., 2018) as well as in several metals and minerals (Tilton, 2014).
In the literature on crude oil, for example, different productivity performance is found between countries using conventional technologies (crude oil from traditional well extraction) and those using unconventional technologies (crude oil from oil sand development, directional drilling and hydraulic fracturing). The difference in productivity performance was associated with the dynamics of technical progress in both types of oil extraction techniques. The former did not show significant variation, with supply concentrated in the producers belonging to the Organization of the Petroleum Exporting Countries (OPEC). However, unconventional techniques have shown productivity gains, especially due to increased research and development (R&D) expenditure (Thuriaux-Aleman et al., 2010). In summary, these two movements have indicated a balanced trend with an almost neutral net effect for the energy extractive industry.
Non-energy mining has undergone a general stagnation since the 1990s. Minerals such as iron, aluminium and copper and coal have shown little improvement in their mining extraction methods. This dynamic of lower technical progress is due to several short-and long-term causes. Among the former explanations are the cyclical effect of lagging investment and lower economic motivations (Tilton, 2014), the declining level of R&D expenditure in previous decades (Radetzki et al., 2008), the lack of specialized workers (Mitchell et al., 2014), decreasing ore grades and depletion of better resources (Mitchell et al., 2014), slower rotation speed of machinery (Mitchell et al., 2014) and higher exploration costs (Koch et al., 2015).
In addition, secular elements have also been pointed out. For Humphreys (2020), it is not easy to measure their impacts as he considers that the new period has three discontinuities that may impact the evolution of productivity. First, there has been a shortage of disruptive innovation for mining with a sectoral specificity different from the new industrial revolution experienced in recent times (Schwab, 2017). Interestingly, this trend is not closely associated with larger scale and bigger machines, which are essential factors for scale economies and productivity. Second, there has been a regular substitution effect between capital/energy and labour in favour of productivity. Although the expectation is that labour costs will continue to rise, the geopolitical scenario is giving rise to unpredictable trends in energy costs. Therefore, the future substitution effect and its productivity impact are not clear. Finally, the less open geopolitical context could affect the flow of international investment and the size of mining projects.
This new situation for the global economy, characterized by greater nationalism and protectionism, is a defining characteristic of BRICS countries and might be related to some particular reasons for cost-push factors. In some cases, the rising cost of exploration and development of new deposits probably has also been associated with new distributive conditions, such as the revival of 'natural resource nationalism' (Serrano, 2013) or the fast growth of real wages in some of them (International Labour Office (ILO), 2013).
Although there is consensus on the slow growth in mining productivity, this fact does not imply that mining extraction remains unchanged or that new methods are not being developed. Beyond the papers on labour productivity, few studies have analysed this other aspect. For example, most of the available articles explored the role of marginal deposits or changes in ore grades at sectoral levels. However, the papers examining the physical side only dealt with depletion trends.
The concepts of ore grades and depletion appear to be similar, but they are very different. Some authors define ore grades as the content of valuable material in extracted material (Priester et al., 2019), whereas others provide broader definitions for depletion. Depletion can include a stripping ratio, harder rocks, more complex mineralogy and more impurities in ore (Humphreys, 2020). In addition, other authors add physical and economic definitions (Rodríguez et al., 2015). Physical depletion has been built on a canonical model that implies a roughly fixed stock and a constant amount extracted each year. Thus, all things being equal, the remaining reserves will be depleted in the next period. Economic depletion is based on the extracting opportunity cost of stock resources (Tilton, 2009). Economic exhaustion exists when production encounters increasing difficulty and eventual non-viability to extract the natural resource due to high mining costs (Rodríguez et al., 2015). In conclusion, the literature could be grouped into two typologies. The first includes authors who see ore grades as a proxy for depletion (Mitra, 2019). The second includes authors who assume a causal relationship between depletion and declining ore grades, a hypothesis that also functions as one of 'sequential depletion' (Rodríguez et al., 2015) and is the one examined in the results section.

METHODOLOGY
The methodology was designed to measure two cost aspects. The first focused on the causal factors of higher prices. Hypotheses were developed by considering three effects. We calculated the impact of annual variation in total labour productivity (TLP) (Pasinetti, 1973) and the pass-through of annual increases in wages and oil import prices. The last two variables were estimated using an input-output cost-push price model (Miller & Blair, 2009). The second aspect accounts for the sequential depletion discussion along with costs (Priester et al., 2019). Thus, we assessed potential cost consequences of switching methods, such as the impacts of increased mining of low-grade ore deposits or increased extraction of non-conventional oil. To analyse this issue, we developed three indicators: compensatory extraction (CE), indirect extraction (IE) and variation in differential rent (r).
Data were drawn from the WIOD 2013 and 2016 releases, and the period of the analysis covers the period 2000-08. Period selection was motivated by the fact that between those years the mining price boom was the longest and broadest since the Second World War. The end year was set due to the reversal of some direct causes, coupled with the financial crisis and the subsequent global economic downturn. In addition, another relevant element was the combination of environmental information restrictions (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009) and the input-output requirements for prices and rent .
Countries were selected on the basis that emerging economies have played a major role in the recent period, in terms of both their demand dynamics and their supply share. Among all emerging economies, BRICS countries such as Brazil, Russia, India and China became crucial because of their size and geopolitical power. Nowadays, they are mining countries with a large impact on the world market and are key for future perspectives. Note that South Africa was excluded because this country does not register any information in WIOD.
Environmental information was required for the ore grade ratio and gross extraction size. The aim of the former is to analyse the efficiency of physical resource extraction, whereas the latter variable separates the impact of the magnitude of extraction from the variations in inefficiency or geological characteristics.
Data were expressed in total tonnes of extraction of mining materials in the 'Mining and Quarrying' sector. The statistical unit was thousands of tonnes, including the use of fossils (coal, gas and oil) and minerals (construction, industry and metal) with and without economic value. The former are materials that enter the economy for direct consumption and as inputs, whereas the latter are materials that will never enter the system, such as physical market externalities. This category includes overburden and parting materials from mining (Genty et al., 2012).

FIRST GROUP OF INDICATORS
Equation (1) refers to total labour productivity (TLP) or 'integral productivity' according to De Juan and Febrero (2000), where TLP i is the total labour productivity of each vertically integrated sector (VIS); v i is the total labour in each VIS (an element in the vector defined as a T l ðI À AÞ À 1 ; a T l is the direct labour coefficient transpose, I is the identity matrix and A is the technical coefficient matrix). As defined above, total labour includes direct and indirect employment, a concept adopted from the VIS sectors proposed by Pasinetti (1973). The first term is the sum of those employed in 'Mining and Quarrying' for each output unit, whereas the second term sums those employed in the sectors selling inputs to 'Mining and Quarrying'. The advantage of its computation is that it considers all the factors and interdependencies and does not include distributional variations (De Juan & Febrero, 2000).
Equation (2) is attributed to vector mining prices after wage increases, P m w :. It is based on the Leontief price model (Dietzenbacher, 1997;Miller & Blair, 2009). Specifically, we focused on taking total differentials (Aydoğuş et al., 2018) and adapting them for our purposes (see the Appendix in the supplemental data online). The indicator includes Δw (real wage increases per year), a T w as a transpose vector of total wages per unit of output and the Leontief inverse matrix. Therefore, if there are wage changes, the indicator has to identify the direct and indirect effects on prices due to inter-industry relations. Computations were registered for all the years included in the research and the cumulative effects are the sum of all the variations until 2008. The real wage increase was calculated with the scalar of uniform wage per country and its percentage changes between periods.
Likewise, equation (3) refers to vector mining prices after oil price increases, P m o :. In this case, computations were done with Δp o (for import oil price increases) and a T m (for the oil import coefficient transpose). Thus, if there is variation in oil importing sectors, its impact on mining and all other sellers' sectors is recognized. Similarly for wages, computations were registered over the eight-year period covered by this study and the cumulative effects are the sum of all variations until 2008. In summary, all computations took a base period (t = n) and a result period after the wage change (t = n + 1). What is presented as cumulative effects is the sum of the eight variations, with n from 2000 to 2008. In each computation, we assumed that the Leontief matrix remains the same as the uniform wage rate varies. Note that there is no impact of prices on the Leontief matrix because no relative prices are present, only a price index.

SECOND GROUP OF INDICATORS
Equation (4) represents the indicator of ore grades in the mining sector. The ratio g is a proxy for the purity or efficiency in extraction activities. The ratio defines the fraction between used material extraction (with economic value) and gross mining extraction (used and unused materials from mining extraction). Specifically, this ratio is the weighted average of the six typologies of ore grades (g i ), where λ i is the extraction share for each typology of fossils and minerals (for further details, see the Appendix in the supplemental data online). This definition of ore grades may of course differ from other studies because it is a sectoral combination of fossils and minerals. Moreover, it is a measure for all mining processes (including feet grades and concentrate grades).
Equation (5) shows indirect extraction variation (ΔIE 2 ) of the mining sector (sector 2) after technical changes, where a 00 e is a ratio between used material extraction and gross output for the year 2000; b is an element of matrix B, which is ðI À AÞ À 1 , the Leontief inverse matrix. For instance, for the year 2000, is the contribution of sector 2 to the total impact on output of sector 2 (initial, direct and indirect) after a unit variation of demand in sector 2. Assuming a constant relationship between physical extraction and gross output, this evolution of total material only changes with variations of the Leontief inverse. Considering only the mining element, its net changes reflect more or less indirect mining extraction inputs, without initial effect (Miller & Blair, 2009, p. 245).
Equation (6) is the rent variation between 2008 and 2000, where P io represents the computable input-output prices (with the observed rate of profits); and P � r is the value of production prices (with a uniform rate of profits). If the differential rate is assumed to be a proxy for rent, the differences between these prices reflect the evolution of differential rent. The same is true for variations in wages, oil prices and productivity; we analysed the cumulative growth of these indicators between 2000 and 2008.
Static input-output prices are regularly used in the traditional measurement of total factor productivity growth (Wolff, 1985). In this study, a modified version of this methodology was used. First, given that not all employment is wage labour, the average money wage rate was computed as the ratio between total wages and total employees. Profits are the gross operating surplus net of hypothetical payments for labour elements. The difference was in the observed profit rate, which was computed for both working and fixed capital and affected the definition of prices in conjunction with the total capital vector.
In addition to these definitions, a modified employment vector was used to normalize wage differentials between sectors. In other words, wage differences between sectors were computed as differences in employment by activity. For example, if an activity pays twice the average, when setting an equal wage, the number of workers must be corrected so that the total spent on wages remains the same. Therefore, in this case it was assumed that the activity pays the average wage and hires twice the number of workers. In short, the employment vector used involves homogeneous labour units and wage differences are related to asymmetries in labour skills and qualifications. This assumption is relevant because it allows associating the bias in prices exclusively to the deviation from the equal profit rate to identify situations in which marginal techniques are used.

RESULTS
Over the past few decades, geopolitical changes have had numerous impacts. It has been universally perceived that natural resources play a central role and that mining and quarrying goods are vital. They are relevant as raw materials for the production of different final goods and services (e.g., lithium for batteries or copper for wire connections) and even for capital goods (e.g., iron ore for steel machinery). Moreover, their centrality is due to their energy supply function, the most fundamental requirement. Therefore, the question as to what kind of relationship exists between the new geopolitics and the mining production conditions arises.
How has this new geopolitical map evolved in global mining at the sectoral level? The following analysis of production is the first step to understand it. Figure 1 plots the recent evolution of the share in value added (VA) and total mining extraction (ME) of all countries included in the database for the BRICS as a whole from 2000 to 2008 ( Figure 1A) and the four selected BRICS countries in 2008 ( Figure 1B). Figure 1A shows a process of growth and concentration for both variables with an increase in the share of the BRICS from 2000 where extraction has always been higher than VA. Specifically, growth was approximately 11.6 and 14.4 percentage points more for VA and ME, respectively, with respect to the global trend. Beyond this rapid rise, these countries also accounted for large shares: these four countries alone accounted by 2008 for one-third of the world's mining value added and almost half of the world's mining extraction.
This general trend concealed some important changes. As Figure 1B shows, in 2008 alone, the shares of the four BRICS countries were not symmetrical with those of Asia. Together India and China accounted for 56% of VA and 80% of the total extraction in 2008, while China accounted alone for 49% of mining VA and 63% of mining ME.
Furthermore, in Asian countries, the share of extraction exceeds the share of value added reflecting the predominance of fossil (e.g., coal) and mineral (e.g., sand used in construction) exploitation, where value is low in relation to the weight of materials.
In contrast, higher value-weight ratios are associated with specialization in industrial metals and oil fossils. This particular typology was observed in the cases of Russia and Brazil.
These new dynamics undoubtedly reflect some of the demand and supply factors discussed above (Hellmer & Ekstrand, 2013). Among the former, population growth and the speed of urbanization and economic development in China and India were the significant elements behind their high production. However, their low productivity (Bartos, 2007) has also been a major driver of rapid development of production in other emerging economies, such as Brazil or Russia.
This evidence concerning the share of emerging economies in the global market is a relevant reason to ask about cost evolutions and their impacts on global trends. For this purpose, Figure 2 plots the cumulative effects on the BRICS mining and quarrying cost from 2000 to 2008 inclusive of productivity, wages and import prices. As mentioned, this period is referred to as the first boom cycle. From 2008 onwards disruptions were caused by the global financial crisis and the emergence of new elements that initiated the second stage of the cycle, which can be seen as a continuation of the first stage and lasted from 2010 to 2012 (Erten & Ocampo, 2013;Gorenstein & Ortiz, 2018). Unfortunately, we did not have full access to the data required for the analysis of these years.
The productivity, oil import prices and wage indicators were selected as they are identified as the main causes of price increases in the literature (Serrano, 2013). However, few studies have measured their quantitative importance in a single methodological framework.
The results show that the mean total increase was 18%. China was the leader (36%), followed by Russia (17%), India (11%) and Brazil (9%). Of the main drivers of these increases, wage rises were the key element. Labour compensation improvements increased prices by almost 18% on average, with China (38%) and Russia (28%) being the main contributors to the trend. The other positive element was the pass-through of oil prices, which averaged approximately 3%. The largest impact came from China (6%), whereas the smallest impact was observed in Russia (1%), which is logical because of the respective roles of domestic production. Both values are compatible with the lack or abundance of natural fossil resources of each country. Besides these two positive effects, improved technology was not relevant for the new price level. Technical progress was only positive in China and Russia reducing prices by 7% and 13%, respectively. In contrast, both India and Brazil recorded negative productivity growth, which had a positive impact on prices (1% and 7%, respectively).
In line with some of the literature, all the selected BRICS countries had a positive impact on costs. The countries with the highest increases were China and Russia, whereas Brazil and India were associated with moderate growth. Overall, the most important determinant of price increases was wages, and the pass-through of oil prices as productivity changes had small impacts on both cost increases and decreases (approximately −3%). Technical progress was low or even negative in the Brazilian case, where the impact on prices was 7%.
A further possible driver of mining cost is sequential depletion, although on this subject no firm conclusions exist (Priester et al., 2019). Some authors consider that lower purity is associated with scale economies. When low ore grades are available, larger amounts of material are needed to mine the same output. A change of this kind can be associated with greater scale economies in particular steps of the chain, such as the drilling or transport stage (Crowson, 2012), but, conversely, lower ore grades can make other stages, such as concentrating and smelting operations, more expensive (Mitra, 2019). Figure 3 reports the impacts of changes in ore grades, rent and indirect extraction between 2000 and 2008 in the four chosen countries.
Three possible paths were identified. First, observation of negative changes in ore grades and positive changes in rent and indirect extraction can be considered as evidence in favour of a 'specific sequential depletion' with a cost impact. In this case, new output is likely to come from marginal deposits, such as remote ones or even from deposits with more complex mineralogy. Recent evidence (Hashiguchi et al., 2021) suggests that this result may reflect higher backward linkages and, therefore, greater resilience to economic shocks. However, the sources of a rise in resilience were left for future analysis of the ratio between services and goods required, which is fundamental to better understand the nature of resilience. Second, lower ore grades and indirect extraction without rent growth are indicative of a process of 'general sequential depletion', increasing the cost of both efficient deposits and marginal areas. Third, it was assumed that 'rent without any depletion' could derive from two sources. It could originate from improvements in efficient techniques or from displacement from efficient to marginal areas due to lack of capacity or for political reasons. Of these possibilities the first was considered only to benefited rentiers, while the second also had an impact on operating costs.
The first case was identified in Brazil. The literature confirms that the Brazilian experience could be linked to a 'specific sequential depletion' derived from the displacement of mining to marginal zones. For example, some authors recognized that new mining developments included the exploitation of non-conventional technologies in energy mining. In particular, ethanol and offshore camps in the pre-salt region were pointed out (Bridgman et al., 2011).
The 'general sequential depletion' was suitable for China. In this case, some authors demonstrated that this process also involved a geographical displacement, from south to north (Roberts et al., 2016;Shen & Gunson, 2006). Although South China is an expensive zone for mining, it is also possible to think of North China as more onerous in terms of transport and exploration cost (Zhang et al., 2015) or even extraction effort (Li, 2018). In the third case, the 'rent without any depletion' profile was identified in Russia and India. In Russia, this case may be associated with a cost impact, as a geographical shift of production was identified. Existing literature has pointed to a geographical displacement in fossil extraction from the Ural/Volga region to the Arctic, East Siberian and Far West districts. The latter are more expensive because of the kind of oil extracted (offshore in the Arctic) and the distance to refining and distribution centres. The causes in the sources of information are not only economic. As Mazat (2013) reported, Russia was keen to find alternative routes to Europe in order to avoid excessive dependence on existing routes and to make progress in its relations with Asian countries. This path could also be linked to improvements in transport, mechanization for offshore production, which is more capital intensive, and positive progress in R&D programmes (Kempener et al., 2014).
In the case of India there is a particular dynamic. Figure 2 indicates no technological improvements and no major process of rent growth (only 2% in cumulative terms from 2000 to 2008). However, evidence of improved ore grades was identified with India needing 9% less gross extraction. As a result, the data did not show strong evidence in favour of cost increases due to depletion.
It can be argued that the data and the indicator do not demonstrate a strict depletion process and its impact on cost. For example, the path of the indicator could reflect a combination of redirection movements (between fossils and minerals) and depletion. To address this issue, the ore grade trend of each type of extraction must be identified. Accordingly, a structural decomposition analysis (Miller & Blair, 2009, pp. 593-602) between the strict depletion effect ('level effect') and the new mining shares ('composition effect') was performed. This analysis indicated that the 'composition effect' was more important than the 'level effect' in all examined BRICS countries. However, some particular trends that are consistent with the previous analysis were identified.
For example, in the Brazilian case a process of 'specific sequential depletion' was found with weaker evidence of 'strict depletion'. This conclusion derives from the fact that the 'level effect' accounted for 30% of the decrease (−0.5 percentage points) in ore grades. The other 70% was explained by the 'composition effect', related to the new mining typology (+1.1 percentage points).
In China there was a 'general depletion process' followed by the highest percentage of 'strict depletion'. In this case, the level effect accounted for 41% of the total negative growth (−1.8 percentage points) and the rest was followed by new mining of less efficient typologies (2.6 percentage points).
The Russian case involved contradictory effects. The negative impact (−88%) was related to the level effect (−0.6 percentage points), which reflected a minimal 'strict depletion' and could be related to the geographical displacement into the Artic zone. The positive impact (+188%) was related to the 'composition effect' (+1.3 percentage points), which showed the importance of other displacements.
The most interesting case was that of India, where there were improvements in its ore grades, as revealed above. However, the decomposition analysis showed that for this country composition effect accounted for the largest share (98%, +4.1 percentage points). Therefore, the process was caused by structural change that led to another type of mining and not by improvements in the deposits.
Focusing on the specific redirection movements between fossils and minerals, Figure 4 depicts the extraction shares of each fossil and mineral in 2000 to 2008 ( Figure 4A) and the evolution of ore grades in the same period ( Figure 4B). Figure 4A shows that the extraction shares differ. China and Brazil are dominant in the extraction of minerals: in 2008 their total shares represented 71% and 89%, respectively. In the same year, fossils in Russia and India accounted for 74% and 68%, respectively.
The major shifts differed between countries. In Brazil, the share of metals increased (+7%) while declines occurred for industrial (−4%) and construction minerals (−3%). In China a negative 'composition effect' was associate with a growing share in industrial (+4%) and metal minerals (+2%) and a reduced share in construction (−3%) and coal (−2%). Russia's positive 'composition effect' was associated with an increase in construction minerals (+3%) and decrease in gas (−2%), whereas India also improved its share of construction minerals (+4%) and reduced its share of coal extraction (−6%).
In the case of ore grades, there are two important aspects to consider: levels and variations. The first item illustrates the effect on sectoral ore grades of redirection between extractions. Brazil's shift from construction to metals did not favour sectoral ore grades due to lower purity in metal extraction. Therefore, the 'composition effect' was relevant because of the displacement of construction mineral deposits by more metalliferous and oil-producing zones. Similarly, the same movement can be seen in China. This 'composition effect' comes from the displacement from fossils and minerals with high ore grades (e.g., coal and construction minerals) to minerals with low (e.g., metals) and decreasing ore grades (e.g., industrial minerals).
Conversely, this negative composition effect was not observed in Russia and India. The process in Russia had a few positive changes in depletion because the country did not experience a major transformation between fossils and minerals. The low upgrading in ore grades could come from the development of construction mineral deposits with high ore grades. The Indian case revealed some improvements due to the displacement of fossils by minerals. The production of fossils with low ore grades (e.g., coal) was replaced by increased exploitation of construction minerals, with a high degree of extraction efficiency.
The 'level effects' of ore grades show minimal variation in all categories from 2000 to 2008, as it is a long-term variable (Mitra, 2019) and only eight years were under consideration. The only strict sequential depletion process was verified in industrial minerals, which revealed a rapid and real deterioration in mining extraction efficiency. Recall that the category of industrial minerals groups together fertilizers, chemicals, saline, and even uranium and thorium minerals. As shown in Figure 4B, this process was mainly verified in China and Brazil. In China the ratio declined from 54% in 2000 to 16% in 2008. In Brazil this ratio also fell from 74% in 2000 to 48% in 2008.
In summary, sectoral statistics can sometimes reflect a sequential depletion as a blend of two processes. Actually, the rise in cost in the period under study was due more to a redirection of exploitation between different ore grades than to a strictly sequential depletion process (Rötzer & Schmidt, 2018). The former was accounted for in the development in the extraction of metal minerals, whereas the latter was only observed in industrial minerals, mainly in Brazil and China.
These last results indicate another key factor involved in understanding the new production cost situation. The biggest producing area in sectoral terms was supplied by low-grade resources and in a context of restricted depletion in industrial minerals. Therefore, the mining discussion about depletion is not relevant to the cost of almost all fossils and minerals in the short term. The real debate should concentrate on the long term and on industrial minerals, which play a key role in recent economic development and were in a situation of overexploitation.

CONCLUSIONS
The new millennium started with a lot of changes in mining resources. The most relevant was the abrupt and sharp increase in the terms of trade for mining. This research has identified some new explanations for these new levels of prices.
One of the main reasons was the displacement of consumption and production from developed countries to emerging economies, mainly Asian countries such as China and India. These countries have high and increasing costs. Several causal factors were at work, such as domestic transport costs, lower ore grades associated with a redirection of mining between fossils and minerals, some depletion process (restricted to industrial minerals), pass-through of imported oil prices, low technical progress and, fundamentally, wage improvements. The consequences of mining transformations are not limited to producing areas, as these changes also influences the development possibilities in other regions.
The 2000 to 2008 period has saw the rise of some emerging net exporters such as Russia and Brazil. In addition the similar factors just mentioned, these countries were also impacted by geographical displacements, such as of oil production to the Arctic for Russia or offshore camps for oil extraction in the pre-salt region of Brazil.
The transformations in mining sectors in 2000-08 saw an increase in the share of expensive countries, with average cost increases of at least 18%. The evidence shows that these new players are of fundamental relevance for future terms of trade, setting a high floor until technological improvements lower costs. Significantly, displacement plays an influential role in shaping volatility and the development possibilities of the other consumer countries with less developed productive structures.

DISCLOSURE STATEMENT
No potential conflict of interest was reported by the author.

FUNDING
This work was supported by Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET, Argentina).