The evolution of economic structure under technological development

Haoran Pan

February 2004

Tyndall Centre for Climate Change Research

Working Paper 48

The evolution of economic structure under technological development
Haoran Pan
Department of Applied Economics, University of Cambridge Sidgwick Avenue, Cambridge CB3 9DE, UK Tel: +44-1223-335287, Fax: +44-1223-335299 E-mail: Tyndall Centre Working Paper No. 48 February 2004


Summary There are obvious gaps between long-term change of economic structure and its principle driving force – technological progress. History has shown the influence of technological progress on economy and current insights in technological development can almost foresee the coming technological waves in next 50 years, but their possible impacts on economy are not yet assessed. The socio-economic system is so complex that its structural change remains intractable. Mainstream economics avoids the problem of structural change by assuming a constant structure in the short and medium terms, but fails to prove itself in long-term projection. On the other hand, non-mainstream economists including historical and evolutionary economists, who have been theoretically challenging mainstream economics’ ideas on economic change, recommend an evolutionary perspective to link changes in production inputs including intermediate and primary inputs, in consumption patterns, and in investment structure with technological progress. In this research we aim to simulate the evolution of economic structure under technological development. The basic unit that we study is technical or production processes, which are regarded to constitute a sector’s production and to represent specific layers of technology, old or new. The phasing out or in of the old or new technical processes changes a sector’s production form, which in turn induces structural change of economy. The relative position of a technical process in a sector is determined by the installed capacity or capital stock required by the process and that the capacity is driven by investment, namely investment in R&D and investment in application of the R&D achievement to production processes. The former depends on both the public and private sectors’ investment behaviour and policy regime. The latter follows the descriptive theory of the technology life cycle, which in turn depends on the investment in R&D. Driven by these two sorts of investment, technologies develop, production processes change their position, and economic structure evolves along the trajectory of technological development. An exercise on Chinese electricity industry shows that the phasing in of nuclear power generation under high investment will greatly change the sector’s form in next 30 years. 1. Introduction Two big issues currently challenging human society are sustainable development and global change. Both of these issues have a long-term dimension, which demands consideration over the next 50-100 years. The current policy regime aimed to support sustainable development and deal with global change will have profound implication for future society, thus the present assessment of the policy regime demands for at least understanding, even if it is impossible to predict, the future. Combating global warming will transform the present fossil-


fuel society into a non-carbon world, what could be the vision of the future world influenced by today’s environmental policy therefore draws policy maker’s top attention. Historical perspective tells us that technological revolution is the fundamental driving force of long-term change in human society. Although oncoming technological change is to some extent foreseeable, human society is still incapable of assessing the potential influence of new technologies on the society. In the past, we have always identified technological influences ex post; even today, at current stage of development of new technologies, we are not yet clear what will be brought about by the technologies to the society. The socio-economic system is so complex that its structural change remains intractable. Mainstream economics regards economic change to be a process of static or dynamic equilibrium, based on the ideas of Newtonian’s mechanicas. It views the economic system as deterministic, controllable, homogenous, reversible, and moving toward equilibrium (Hall, 1994; Janssen, 1998). It avoids the problem of structural change by assuming a constant structure in the short and medium terms, but fails to prove itself in long-term projection. Its mechanical extrapolation, often based on capital stock, into the future receives most criticisms from non-mainstream economists including historical and evolutionary economists, who have been theoretically challenging mainstream economics’ ideas on economic change by arguing that economic system is undeterministic, uncontrollable, heterogeneous, irreversible, and a system in disequilibrium. They recommend an evolutionary perspective to link changes in production inputs including intermediate and primary inputs, in consumption patterns, and in investment structure with technological progress (Van den Bergh and Gowdy, 2000). In this research we aim to simulate the evolution of economic structure under technological development1. The basic unit that we study is technical or production processes, which are regarded to constitute a sector’s production and to represent specific layers of technology, old or new. The phasing out or in of the old or new technical processes changes a sector’s production form, which in turn induces structural change of economy. The relative position of a technical process in a sector is determined by the installed capacity or capital stock required by the process and that the capacity is investment-driven. We distinguish between two types of investment in technology: namely investment in R&D and investment in application of the R&D achievement to production processes. The former depends on both the public and private sectors’ investment behaviour and policy regime. The latter follows the descriptive theory of the technology life cycle, which in turn depends on the investment in R&D. Driven by these two sorts of investment, technologies develop, production processes change their position, and economic structure evolves along the trajectory of technological development. Thus, the method in this research entails an endogenous Input-output structural change, which likes an evolution mechanism or Schumpeterian creative destruction.


The method that we proposed in this research can be used to study particular policy issues such as climate change and sustainable energy. If a new technology develops towards a direction to lower fossil-fuel use, it is favourable to both resource reservation and climate change, the gross costs combating global warming could be much lower than is naively expected with simplifying treatment of technology. Typically, China now is one of the big coal consumers and its nuclear power accounts only a share of 1.4% of total national power generation. Taking the electricity industry in China as an example, we illustratively project the change of input structure in the electricity sector in China under investment in nuclear power generation technology. An extensive review on Input-output study of structural change issue will be presented in Section 2 and another review on modelling technological change in Section 3. Section 4 will focus on the method in this research. Second 5 will apply the method to China’s electricity industry. Finally, Section 6 concludes this research. 2. Input-output analysis on structural change Since its birth half a century ago, the Input-output table has been widely used in the description of economic structure, due to its explicit presentation of inter-industry relations. Apart from Input-output model, many detailed, disaggregate macro models such as CGE, integrated assessment, and macroeconometric model also employ the intermediate structure of Input-output table at their core. An essential feature of Input-output analysis is the technical presentation of sectoral production, with an assumption of fixed input coefficients. Intuitively, it seems that Inputoutput coefficients will change with technological progress and reflect this progress. More specific, there should be some decline in intermediate and labour inputs as advanced, more productive technologies are phased in the production process. However, evidences commonly show that the Input-output coefficients are quite stable in short term; And even in the medium term, they do not change in accordance with technological progress (Carter, 1970). This paradox makes it difficult for economists and technology analysts to relate Input-output coefficients to technological change directly. The reason might be that the Input-output coefficients result from the effects of various factors, which offset each other in most cases. Therefore, some Input-output economists conclude that coefficient change is by no means the same thing as technological change (Miernyk, 1977). In spite of the observed ambiguity between Input-output structure and technological change in short and medium term, it is strongly believed that Input-output structure should change in accordance with technological progress in long term. This is for two reasons that technological change is a long-term matter and that the highly aggregated Input-output


structure cannot effectively indicate structural change in not long term. The belief that technological change is the fundamental driving force for economic structural change comes partly from theoretical insights and partly from the historical perspective - that in the past technological change has indeed induced the change of economic structure. Since Inputoutput structure is still regarded as one of the best tools in the description of structural change, in order to study structural change under technological change, one has no alternative but to dig in deeper. That is a methodological issue, which has two aspects. One is how to change Input-output coefficients to reflect technological change. Another is that how technological change is described in relation to Input-output structure. Both of the subjects have drawn many studies in the past. Input-output economists have devoted a lot of effort to adjust the fixed Input-output coefficients to capture technological change. The traditional method in adjusting the coefficients was dominated by mechanical adjustment. Among this family, the RAS method (Bacharach, 1970; Allen and Lecomber, 1975) has gained popularity in adjusting and balancing Input-output coefficients, for its ability to be operationalised. In projecting Inputoutput coefficient change across time periods, the method relies on forecasting total inputs and outputs in sectors. Once these numbers are estimated, Input-output coefficients are adjusted proportionally by the ratio of future to present input and output. However, the central problem with the RAS method is its assumption that Input-output coefficients change uniformly along rows and columns. There is no economic justification for changing Inputoutput coefficients in that way (Miernyk, 1977). Another drawback of the method is its reliance on forecasting, which often entails many errors during estimation. An alternative way to reduce forecasting errors is to use an ex ante method, based on experts’ insights into future patterns of technical change in production. However, the ex ante method comes out no better than the RAS method (Almon et al, 1974). Some analyses (Arrow & Hoffenberg, 1959; Forssell, 1970; etc.) focus on the study of the time-series behaviour of Input-output coefficients and attempt to predict long-term trend of change. In practice, the method has hardly been used because it relies heavily on both historical data and forecast explanatory variables and involves operational difficulty. The RAS and time series methods for projection of technical coefficient change are limited in description of technological change for a number of reasons. First, the methods generally do not consider technology explicitly. Technology acts only as a name of a factor explaining the changes over time. The methods therefore cannot separate potential influences of different technologies. Historical data come from the variation in past periods, which include all possible influencing factors, principle (technological change) and trial (relative price change, business cycle, data error). Different technologies were mixed and their stages of


development are ignored. Second, the methods lack an analysis of new technologies and therefore have little to say about the role of new technology in influencing the economy and technological evolution. Finally, the classical methods isolate the change of Input-output technical coefficients from economic behaviour, say investment in innovation and diffusion, and seem incapable of long-term projection. It is Carter (1970) who has explicitly incorporated the new technology into Input-output coefficient change. In the research, each sector consists of an old and a new technological layer. Each coefficient of future flow is defined as the weighted average of an old-technology and a new-technology coefficient. Given the weight as the share of expenditures on new capital assets in gross capital assets, the new-technology coefficient is derived from the future-technology and old-technology coefficients. Obviously, the method is not suitable for the projection of Input-output coefficients, since it assumes that future coefficients are known. Furthermore, the method employs linear programming to determine investment behaviour between old and new technology, without any consideration over technological development itself. Economists often consider the problem with Input-output coefficient change to be the fixed coefficient assumption, which allows no substitution among primary and intermediate inputs, and therefore attempt to introduce flexible production functions such as Cobb-Douglas and CES function to allow substitution among inputs. Johanson (1960) used Cobb-Douglas function for labour and capital inputs, keeping intermediate inputs constant. The method now is widely used in CGE models. However, two problems remain with the application of flexible production functions. First, it is almost impossible to directly include intermediate inputs in such functions. Uzawa (1962) and MacFadden (1963) showed that in this case all inputs have same substitution elasticity. The second problem regards the dynamics of substitution elasticity. How the elasticity changes over time is unknown. In order to avoid the substitution problem with Input-output coefficients, Bernt and Jorgenson (1973) and Jorgenson (1998) completely discarded the Input-output method and instead proposed a translog price function to endogenize economic structural variables. The translog function allows substitution among all factors and is determined by relative price change and technological change that is represented by a time difference. Problems with the method are that it is based on a competitive equilibrium assumption and simplifies technological change to a factor differing over time periods, and its empirical results show that all coefficients are similar (Jorgenson and Fraumeni, 1981). The method also depends heavily on time-series data and, therefore is not widely used.


Leontief and Duchin (1986) were the first who explicitly explored the influence on Inputoutput technical coefficients of information technology in the US economy. They designed both qualitatively and quantitatively four scenarios tracing the possible paths of influence of information technology for the period 1980-2000. Through a dynamic Input-output system, they assessed the influences on the economy of information technology. While their research did capture the evolution of information technology to some extent, it did not relate the information technological change with R&D investment and consider other potential technologies. In other words, technological change is exogenous in their research. With the current popularity of the endogenous growth theory, there are some attempts to incorporate the theory into the adjustment of Input-output technical coefficients. The essence of the theory is that R&D investment will drive technological progress and therefore sustain economic growth. However, the application of endogenous growth theory to Input-output structure is often oversimplified and sometimes misleading. Without explicit description of technological aspects, Los (2001) simply assumes that technological progress is purely labour-saving, to some extent to avoid the difficulty in intermediate input substitution. The treatment assumes technological progress to be an identical, infinite procedure solely shaped by past R&D investment and is even more abstract than classical methods. As a result, the hypothetical empirics in the research rapidly lead to the vanishing of labour from production and exhibit a false ability to predict the pattern of Input-output technical coefficients for infinite periods. 3. Modelling long-term technological change Both the demand for understanding of long-term economic change and the dissatisfaction with the neoclassical economics’ approach on the change are generating a renewed interest in long wave theory, a subject that describes economic change with long-term Kondratiev waves driven by Schumpeterian radical technological change (Freeman and Soete, 1997). Neoclassical economics acknowledges the fundamental role of technological change in economic growth. The classical exogenous growth theory (Solow, 1956) identifies that longterm economic growth will be sustained by exogenous technological progress. The recent popular endogenous growth theory (Romer, 1986) attributes long-term economic growth to the positive externality of cumulative knowledge, which is enhanced by R&D investment within the economic system. It is widely noted that neoclassical approaches, both exogenous and endogenous, on technological change are over-simplistic, ignoring a comprehensive analysis of technological characteristics. This is reflected in a number of applied economic models based on such theories. The exogenous growth economic models commonly assume that technological


change is a deterministic time trend with exponential or whatever form of exogenous factor growth. As an updating version to endogenous growth theory, some formerly exogenous growth models use cumulative knowledge or capacity to represent technological progress to claim itself as an endogenous growth model. For example, Buonanno et al (2000) adds the stock of knowledge in each production function and relates the stock of knowledge to R&D investments. There has been a movement including historical and evolutionary economics and some new thinkers to criticise neoclassical economic theory on economic growth in relation to technological change. Its representatives include Freeman and Soete (1997), Freeman and Louçã (2001), Perez (1983), Nelson and Winter (1974, 1982), Dosi (1988), Ayres (1998), and Janssen (1998). They argue that mainstream economics’ ideas are in accordance with Newtonian mechanics, regarding the economic system as deterministic, controllable, homogenous, reversible, and a process moving from static to dynamic equilibrium. The neoclassical approach on long-term economic change fails to prove itself in practice. Its mechanical extrapolation, often based on capital stock, into the future is unconvincing. Janssen (1998) points out that economic system is undeterministic, uncontrollable, heterogeneous, irreversible, a disequilibrium, evolutionary process. Long-term technological change has not been properly modeled so far, even though some works existed on basis of the “evolutionary” approach. These often use the mathematics developed for dynamic processes in biology (see, for example, Silverberg, 1988 and 1997). However, there is a good descriptive theory to provide insights into technological change. This is the theory on the life cycle of a technology. In the tradition of Schumpeter, who distinguished three important phases in technological development –invention, innovation and diffusion, Freeman and Louçã (2001) further elaborate the life cycle of a technology by six phases as shown below. (1) the laboratory-invention phase, with early prototypes, patents, small scale demonstrations and early applications; (2) decisive demonstrations of technical and commercial feasibility, with widespread potential applications; (3) explosive take-off and turbulent growth, characterized by heavy investment and many business startups and failures; (4) Continued high growth, as the new technology system becomes the defining characteristic of the economy;


(5) slowdown, as the technology is challenged by new technologies, leading to the next crisis of structural adjustment; (6) Maturity, leading to a (smaller) continuing role of the technology in the economy or slow disappearance. Figure 1 (Köhler, 2003) shows the six phases, which forms the S shape. Grübler (1998) points out that the analyses on the development of many technologies confirm the S shape description. Since Griliches (1957), who described the procedure of technological change by a logistic function in an analysis of American agriculture, the logistic curve for the technology life cycle model has been widely used, even though it does not explain why and how the curve is shaped. Figure 1. Phases in the life cycle of a technology and the logistic curve fitting




1 Phase

2 · Year 0


Year 70

The development of a technology and its influence on economic system are yet to be revealed comprehensively. Besides the evolutionary approach, there are some efforts in the examination of interindustry R&D spillover. The penetration of a new technology generated from a certain firm or sector into other firms or sectors mainly comes from the technologyembodied intermediate inputs and capital goods, the so-called rent spillover, and some from the externality of, the knowledge spillover (Griliches, 1979). The knowledge spillover may be significant among firms in sector, but insignificant at sector level. Current analysis of interindustry R&D spillover is far from a comprehensive assessment (Mohnen, 1997). 4. The method 4.1 Specifications The method in this research stays in the framework of Input-output analysis that each sectoral input structure is specified by fixed coefficients of intermediate and primary inputs, but we go


further in the sectoral inputs to look at the production processes within the sector. We regard that each production process represents a sort of technological layer and sectoral input structure is the mixture of all production or technical processes in the sector. The separation of different technical processes will allow explicit consideration over specific new technologies. The structure of intermediate and primary inputs of a production or technical process is the technical structure of the process, which is assumed fixed throughout time by its nature. The dynamics of a sectoral input structure depends on the phasing in and out of new and old technical processes. This means that substitutions can take place among primary and intermediate inputs in a sector. The method further defines that the relative position of a technical process in a sector is determined by the installed capacity required by the process and that the capacity is investment-driven. We distinguish between two types of investment in technology: namely investment in R&D and investment in application of the R&D achievement to production processes. The former depends on both the public and private sectors’ investment behaviour and policy regime. The latter follows the descriptive theory of the technology life cycle, which in turn depends on the investment in R&D. Driven by these two sorts of investment, technologies develop, production processes change their position, and economic structure evolves along the trajectory of technological development. Thus, the method in this research entails an endogenous Input-output structural change, which likes an evolution mechanism or Schumpeterian creative destruction. Figure 2 shows the evolutionary process of sectoral production process. In period 1 the sector consists of ten old technical processes (the hexagons), but in period 2 a new technical process (the pentagon) emerges. The new process, driven by investment, will gradually expand its share in the sector production and kick out the old processes. In period 8, a newest process (the diamond) appears and will increase its share in the sector during following periods. It is clear that in the last period in Figure 2 the old technical processes will be eliminated out of the sector, the new technical process will dominate the sector production, and the newest technical process will compete with the dominator. As a result, the sector will completely transform its production techniques.


Figure 2. The evolution of sectoral production process

Sectoral productivity

Focusing on the importance of investment and capacity building in changing economic structure, we assume that labour demands can always be met through training and education. It is important to notice that this method is also able to address the productivity growth enhanced by technological progress through the phasing in of new technological processes. To project the Input-output coefficients, it is desirable to have the Input-output data in terms of each new technical process. However, existing statistics has not yet provided with such detailed and updated data. The construction of such new technological Input-output tables may beyond our ability and resources, we thus adopt a calibrating method to access the information on new technical process. 4.2 The projection method Denote intermediate input coefficient matrix as A, the elements of which are assume consists of an ordinary technical process with technical coefficient technical process with technical coefficient product in sector j’s ordinary process and and

a ij , which we
O aij and a new

N O aij . Let xij indicate the input of sector i’s

N xij in sector j’s new technical process, and X O j

X jN are sector j’s total output from ordinary and new technical process respectively. Let
indicate the price of sector i’s product,


Pi O and Pi N

are the prices of ordinary and new

production process, respectively. At base year 0, we have the following coefficients,


For the ordinary technical process,
O aij = O Pi ,O xij ,0 0

PjO0 X O,0 , j


For the new technical process

a =
N ij

N Pi ,N xij ,0 0

PjN0 X jN,0 ,
O N Pi ,O xij ,0 + Pi ,N xij ,0 0 0


The coefficient from Input-output table is

aij , 0 = =a ⋅
O ij

PjO0 X O,0 + PjN0 X jN,0 j , , PjO0 X O,0 j , PjO0 X O,0 + PjN0 X jN,0 j , , +a ⋅
N ij

PjN0 X jN,0 , PjO0 X O,0 + PjN0 X jN, 0 j , ,



Above says that the combined Input-output coefficient is the average of coefficients of ordinary and new technical processes, weighted by the share of the product of each process in total product. Apart from intermediate inputs, the operation of the new technical process also requires labour and capacity input, which we regard as new-technology-specific as well. The newtechnology-specific labour is generated through training and education and the newtechnology-specific capacity is accumulated through investment in the new technology. Assuming new-technology-specific labour supply can easily meet demand through intensive training and education, the expansion of the new technical process in fact will be only constrained by new-technology-specific capacity. In other words, we assume full utilisation of new-technology-specific capacity. Denote

k jN as the coefficient of new-technology-specific

capacity use in sector j, at base year 0, we have

k =
N j

rjN0 K jN,0 , PjN0 X jN,0 ,


Where capital K j indicates new-technology-specific capacity or capital stock built in sector j and


γ N the rental rate of the capacity or capital stock. Similarly, the coefficient of ordinary j

capacity use in sector j is

kO = j

rjO0 K O,0 j , PjO0 X O,0 j ,

And, the coefficient of the combined capacity use as required by production is

k j ,0 = k ⋅
O j

PjO0 X O, 0 j , PjO0 X O,0 + PjN0 X jN,0 j , ,

+k ⋅
N j

PjN0 X jN,0 , PjO0 X O,0 + PjN0 X jN,0 , j ,



Assuming the capacity in each sector is fully utilised, the output from new technical process can be derived,
N j ,0 N j ,0



γ N, 0 K jN,0 j
k jN,0


And for ordinary process


O j ,0

O j ,0


γ O,0 K O,0 j j
k O,0 j


Substituting (7) and (8) into (3) and (6) respectively, the combined coefficient of intermediate use becomes dependent to capacity

aij , 0 = a ⋅
O ij

k jN K O,0 j k jN K O, 0 + k O K jN,0 j j

+a ⋅
N ij

k O K jN,0 j k jN K O, 0 + k O K jN,0 j j



And, the combined coefficient of capacity use as well

k j ,0 = k O ⋅ j =

k jN K O,0 j k K
N j O j ,0

+k K
O j

N j ,0

+ k jN ⋅

k O K jN,0 j k jN K O,0 + k O K jN,0 j j

k O k jN ⋅ ( K O,0 + K jN, 0 ) j j k jN K O, 0 + k O K jN,0 j j



The formula (9) and (10) show that a combined Input-output coefficient is dependent to the input structure and capacity of each process. If each process’s input structure is fixed, the combined Input-output coefficient changes with capacity change in each process. We further discuss investment and capacity accumulation below. Denote δ N as depreciation rate and I as investment; we accumulate the new-technologyspecific capacity in a classical way

K jN,1 = (1 − δ N ) K jN, 0 + I jN,1 ,
depreciation in period 0. Similarly, the capacity in ordinary process is accumulated as


which says that the capacity at period 1 is the investment in the period plus the capacity net of

K O,1 = (1 − δ O ) K O, 0 + I O,1 j j j
different capacity. In sector j, total investment is the sum of the two sorts of investment


Note that we use different depreciation rates in different equations, as the two processes have

Ij = IO + IN . j j


We further define the investment in new technology application dependent on new technology development, which will go along its life cycle - a logistic-like curve. In this research, the logistic curve is not fixed; how fast a new technology will go through the cycle depends both on the nature of the technology and on the R&D investment in the technology. As Figure 3 shows, if a technology will go through a life of 60 years with a normal level of investment, additional investment to the level will accelerate the development of the technology and shorten its life span to, e.g. 30 year. This perspective is for example particularly important in assessing induced technological change. Heavy investment in environmental technology will rapidly develop the technology, which in turn will assist in solving environmental problems.


Figure 3. The dependence on R&D investment of the life span of a technology

30 years Heavy investment

60 years Moderate investment

90 years Light investment


I jN = θ j ⋅ Dt .


Where θ j is parameter vector and Dt the indicator of technological performance, which can be output, market share, inverse of cost or price, or something else, following a logistic curve with a definite starting year but a varying ending year depending on the R&D investment in new technology. That is

Dt =

1 1 + e −α t + β


Where β determines the lower tail span. In terms of technology development, it describes the process of phase 1 and 2 and defines their spans. We empirically specify its value. Here


is a variable, which defines the logistic curve’s span from very low to top. In terms of technology development, it describes how fast a technology will reach its peak of life. We relate this variable to R&D investment, assuming that additional investment will accelerate a technology’s development and therefore shorten its life cycle. Denote

I tR & D as index of R&D

investment in new technology R&D sector; we assume a linear relationship between the index and


α t = τ ⋅ I tR & D



Again, parameter

τ can

be empirically determined. In a multinational context, the R&D

investment includes both domestic and foreign investment and therefore reflects spillover effect across nations. We will approach the R&D investment behaviour in an integrated macroeconomic system, which currently is beyond the scope of this paper. Finally, for period 1, we obtain the new coefficient of intermediate use

aij ,1 = a ⋅
O ij

k jN K O,1 j k jN K O,1 + k O K jN,1 j j

+a ⋅
N ij

k O K N,1 j j k jN K O,1 + k O K N,1 j j j



and the new coefficient of capacity use

k j ,1 =

k O k jN ⋅ ( K O,1 + K jN,1 ) j j k jN K O,1 + k O K jN,1 j j



From the equation (17) and (18), it is clear that both the new coefficients of intermediate and capacity use are investment-driven, given the fixed technical structure of production processes. However, because of its aggregating nature, the existing Input-output table cannot provide input structure of each production process, the data only represents a mixture of all different processes, therefore the technical structure of a single process is not observable. Considering the difficulty in collecting the data, we calibrate the technical structure of the ordinary and new technical processes in this study, using time-series Input-output and capacity data. 4.3 The calibration of the technical structure Assuming a new technical process has not entered production in base year 0 and other technologies remain constant from period 0 to 1, we then can take
O aij = aij ,0


k O = k j ,0 , where both aij ,0 j


k j ,0

are observed data. If the new technical process

appears in period 1, then from equation (18) the coefficient of capacity use of the new technical process can be calibrated as below where the RHS of the equation (19) all are known variables,

k jN =

k j ,1k O K jN,1 j k O ( K O,1 + K jN,1 ) − k j ,1 K O,1 j j j

Subsequently from the equation (17), we can calibrate the coefficient of intermediate use of the new technical process


a =(
N ij

k jN K O,1 j k O K jN,1 j

+ 1) ⋅ (aij ,1 − a ⋅
O ij

k jN K O,1 j k jN K O,1 + k O K jN,1 j j



Once the technical structure of a new technical process is calibrated, we are ready to project Input-output coefficients to future. However, it is worthy to notice that the assumption that other technologies except for the new technology in question are fixed over periods is not really operational, as the data always shows a mixture of all technologies’ change. This problem, which we cannot solve at present, may adversely affect the quality of the calibration. 5. The change of China’s electricity industry under nuclear power investment According to the method above, we can project Input-output coefficients under various technologies such as information technology, biotechnology, nano-technology, renewable energy technologies, nuclear power technology, and so on. The projections of these technologies may be complex or simple, depending on the technical context and development stage of the technologies. For example, information technology has been applied in all sectors, thus has impacts on the production structure of all the sectors. On the other extreme, nuclear power generation technology may be one of the simplest to examine its impact on economic structure, because the technology can be mainly used for base-load power generation within electricity industry. More importantly, the nuclear power technology plays a key role in sustaining world energy and provides a benign solution for the world to drift away from fossil-fuel use. As Figure 4 shows, China’s current power generation relies mainly on coal, the coal used in electricity industry accounts over 50% of total coal consumption, thus studying the impact of nuclear power technology on Chinese electricity industry has particular implication in China’s reform on energy market.
Figure 4 Share of coal used for power generation in total coal consumption in China
0.70 0.60 0.50 Share 0.40 0.30 0.20 0.10 0.00

19 89

19 87

19 91

19 93

19 73

19 71

19 75

19 77

19 85

19 79

19 81

19 83



20 01

19 95

19 97

19 99

Source: IEA (2003), Energy Balances of Non-OECD Countries 1971-2001. China started building its first nuclear power plant, the Qinshan, in 1985. The plant operated in 1994 and since then China has constructed total eight nuclear power operators, with total net capacity at about 6000 Mwe per year, a share of 1.4% in total national power generation. It is worthy to notice that most of the constructed operators was mainly relying on foreign technology and investment. In recent years, China has developed its own nuclear power technology into a sufficient stage and is planning to construct two operators, each at 1000 Mwe, per year. This may indicate that the nuclear power technology in China now is in its high growth phase, the fourth phase of technological life cycle as analyzed in Section 3. Figure 5 a-c show the scenarios of investment in construction of nuclear power plant when R&D index is 1.05, 1.2 and 1.35 respectively. It is clear that if R&D index is 1.05 it will take more than 50 years for the nuclear technology to be matured in China; if R&D index is 1.35 the procedure may be ten years shorter.


Figure 5a Investment in new technology when R&D index is 1.05

1.200 1.000 0.800 0.600 0.400 0.200 0.000 0 10 20 30 Year 40 50 60 70


Figure 5b Investment in new technology when R&D index is 1.2
1.200 Investment 1.000 0.800 0.600 0.400 0.200 0.000 0 10 20 30 Year 40 50 60 70

Figure 5c Investment in new technology when R&D index is 1.35
1.200 Investment 1.000 0.800 0.600 0.400 0.200 0.000 0 10 20 30 Year 40 50 60 70

To trace the application of nuclear power technology in China, we use the 1987 and 1997 Input-output tables for China. (Both tables are deflated to 1990 price). We regard the 1987


table containing no nuclear power technology and the 1997 table including significant nuclear power technology. Table 1 gives the input structures of the electricity sector in these Inputoutput tables. In this exercise, we take the index of R&D investment as 1.35 and assume the investment started from the year 1985. As shown in Figure 5c, the 1987 Input-output table described a stage almost without installation of the capacity for nuclear power generation, whereas the 1997 Input-output table indicates some new capacity installed. According to the calibration method in Section 4.3, we calibrated the input structure of the nuclear power production process within electricity sector, as shown in Table 1. Two problems emerge when the calibrating results are examined. Firstly, some small negative numbers appear in the input list. We analyze that this is subject to the investment level that we specified. Intuitively, the new process will require no coal for production rather than produce coal. We thus force all the negative numbers to zero, which means that the nuclear power production will not require the relevant inputs. Secondly, we cannot exclude the possibility that the calibrated results also embody the impact of other technologies. As the result suggests for the sector of agricultural and industrial machines, the new process will demand over 15% of total input from the sector. In spite of lacking data verification, we guess that it is very likely that hydropower or others drive the strong demand for the machines. With respect to the sectors of office machines, electrical goods, and communications, where the new process’ demands are also very strong, we analyze that information technological development plays important role in driving the demands. Because of data and resource limitation with this research, we cannot separate the electricity industry in terms of each of different affecting factors. We therefore assume that 90% of the demands for the products of office machines, electrical goods, and communications is due to information technological development and 70% of the demands for the products of other sectors come from hydropower or other power installation. Similar story happens to capital input coefficient, which decreases from 0.5 in 1987 to 0.35 in 1997. Obviously, we cannot attribute all the change to nuclear power’s installation and therefore assume that only 50% of the change are due to nuclear power. Using these assumptions, we finally obtain the new process of nuclear power generation, which is then used to project the Input-output coefficients for the period from 2002 to 2032. In the year 2032, the development of the nuclear power technology will be saturated in China and the electricity sector will only require 0.0147 Yuan coal for one Yuan electricity generation by 1990 price, which just accounts to about 12% of the figure in 1997 Input-output table.


6. Conclusion This paper has attempted to connect specific technologies, R&D investment, capacity building, and Input-output coefficient all together to simulate sectoral evolution in long term. R&D investment drives the development of new technologies, capacity building brings new technical processes into sectoral production, technical process within a sector exchange their relative positions, and sectors change their forms along time. The method proposed in this paper critically depends on identification of the new technical process. It is ideal to use an Input-output table in terms of a specific technology, but the data unfortunately is unavailable at present or doesn’t exist at all. The second best way is to use engineering data to specify the input structure of nuclear power generation. Instead, in this research we calibrated the new technical process, using time-series Input-output tables and historical data on investment. The quality of the calibration is subject to a number of factors but enough for illustrative purpose. In the small exercise with respect to Chinese electricity industry and nuclear power generation, the analysis focuses on illustrating the use of the method rather than simulating the process in which nuclear power phases into Chinese electricity market. In later research, we will conduct the simulation under a series of scenarios regarding investment behavior and policy concerns. Acknowledgements This research was conducted within the ETEC project under the funding support from the Tyndall Centre for Climate Change. I am grateful to the Centre and the following colleagues, within the Etech+ project, who provided with valuable suggestions and ideas: Paolo Agnolucci, Dennis Anderson, Terry Barker, Paul, Dewick, Paul Ekins, Tim Foxon, Ken Green, Robin Hankin, Jonathan Köhler, Marcela Miozzo, Rachel Warren, and Sarah Winne. Pierre Mohnen and Thijs ten Raa in particular are thanked for their comments on an early draft.


Table 1 Projection of Input-output coefficients under nuclear power technology
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Electricity sector Agticulture etc Coal & Coke Oil & Gas Extraction Gas Distribution Refined Oil Electricity etc Water Supply Ferrous & Non-F Metal Non-metaliic Min. Pr. Chemicals Metal Products Agri. & Indust. Mach. Office Machines Electrical Goods Transport Equipment Food, Drink, Tobacco Tex., Cloth. & Footw. Paper & Printing Pr. Rubber & Plastic Pr. Recycling/Ems Abatement Other Manufactures Construction Distribution etc. Lodging & Catering Inland Transport Sea & Air Transport Other Transport Communications Bank, Finance & Ins. Other Market Serv. 1987 0.000550 0.184195 0.042810 0.002453 0.057755 0.016679 0.005838 0.007378 0.007754 0.006745 0.004913 0.016214 0.006507 0.015469 0.002889 0.000904 0.007032 0.002553 0.003004 0.000000 0.005904 0.000000 0.025004 0.000458 0.018824 0.002088 0.000718 0.000197 0.002243 0.000677 1997 New process 0.000386 -0.000527 0.120274 -0.235962 0.021759 -0.095559 0.000548 -0.010068 0.022970 -0.170890 0.021258 0.046773 0.003302 -0.010834 0.004068 -0.014376 0.008293 0.011295 0.006189 0.003094 0.004366 0.001323 0.057378 0.286788 0.028701 0.152390 0.056608 0.285875 0.007707 0.034555 0.000000 -0.005040 0.004947 -0.006673 0.000215 -0.012816 0.002680 0.000874 0.000000 0.000000 0.009947 0.032478 0.002860 0.018796 0.065409 0.290589 0.002536 0.014114 0.031974 0.105264 0.001985 0.001412 0.000002 -0.003990 0.002416 0.014788 0.022979 0.138543 0.014271 0.090032 Adjustment 0.000000 0.000000 0.000000 0.000000 0.000000 0.014032 0.000000 0.000000 0.003389 0.000928 0.000397 0.086036 0.015239 0.028588 0.010367 0.000000 0.000000 0.000000 0.000262 0.000000 0.009744 0.005639 0.087177 0.004234 0.031579 0.000424 0.000000 0.001479 0.041563 0.027010 2007 0.000162 0.054241 0.012607 0.000722 0.017007 0.014811 0.001719 0.002173 0.004674 0.002641 0.001727 0.065475 0.012667 0.024725 0.008165 0.000266 0.002071 0.000752 0.001069 0.000000 0.008613 0.003978 0.068868 0.003122 0.027823 0.000914 0.000211 0.001101 0.029984 0.019255 2017 0.000059 0.019917 0.004629 0.000265 0.006245 0.014318 0.000631 0.000798 0.003861 0.001557 0.000885 0.078486 0.014295 0.027169 0.009558 0.000098 0.000760 0.000276 0.000559 0.000000 0.009328 0.005029 0.080454 0.003826 0.030200 0.000604 0.000078 0.001340 0.037311 0.024162 2027 0.000045 0.015133 0.003517 0.000201 0.004745 0.014249 0.000480 0.000606 0.003747 0.001406 0.000768 0.080300 0.014522 0.027510 0.009752 0.000074 0.000578 0.000210 0.000487 0.000000 0.009428 0.005175 0.082069 0.003924 0.030531 0.000560 0.000059 0.001373 0.038332 0.024846 2032 0.000044 0.014703 0.003417 0.000196 0.004610 0.014243 0.000466 0.000589 0.003737 0.001393 0.000757 0.080463 0.014542 0.027540 0.009770 0.000072 0.000561 0.000204 0.000481 0.000000 0.009437 0.005189 0.082214 0.003933 0.030561 0.000557 0.000057 0.001376 0.038424 0.024908


31 32

Non-Market Services Un-allocated Total intermediate inputs capital input Labour input Total input

0.004340 0.000000 0.452095 0.496997 0.050908 1.000000

0.004037 0.000000 0.530065 0.423293 0.046642 1.000000

0.002349 0.000000 0.964596

0.000705 0.000000 0.368789 0.012531 0.618680 1.000000

0.001775 0.000000 0.393320 0.155194 0.451485 1.000000

0.001098 0.000000 0.377797 0.064917 0.557286 1.000000

0.001003 0.000000 0.375633 0.052334 0.572033 1.000000

0.000995 0.000000 0.375439 0.051202 0.573359 1.000000


References Allen, R. and J. Lecomber (1975), ‘Some Tests on a Generalized Version of RAS’, in R. Allen and W. Gossling eds., Estimating and Projecting Input-Output Coefficients, Input-Output Pbulishing Co., London. Almon, C., B. Buckler, L. Horwitz and T. Reimbold (1974), Interindustry Forecasts of the American Economy, D.C. Heath and Co., Lexington Books, Lexington, Massachusetts. Arrow, K. J. and M. Hoffenberg (1959), A time series analysis of interindustry demands, North-Holland, Amsterdam. Ayres, R. U. (1998), ‘Towards a Disequilibrium Theory of Endogenous Economic Growth’, Environmental and Resource Economics 11: 289-300. Bacharach, M. (1970), Biproportional matrices and Input-output change, Cambridge University Press. Berndt, E. and D. Jorgenson (1973), ‘Production Structure’ in D. Jorgenson and H. Houthakker eds., U.S. Energy Resources and Economic Growth, chapter 3, Washington, DC: Energy Policy Project. Buonanno, P., C, Carraro and M. Galeotti (2000), ‘Endogenous Induced Technical Change and the Costs of Kyoto’, Workshop on Economic Modelling of Environmental Policy and Endeogenous Technological Change, Amsterdam. Carter, A. (1970), Structural change in the American economy, Harvard University Press. Dosi, G. et al., eds., Technical Change and Economic Theory, Printer Publishers Ltd., London. Forssell, O. (1970), ‘Explaining changes in Input-output coefficients for Finland’, in Carter, A. P. and A. Brody (eds), Applications of Input-output Analysis, NorthHolland, Amsterdam. Freeman C. and F. Louçã, 2001, As Time Goes By, Oxford: OUP.


Griliches, Z. (1957), ‘Hybrid corn: an exploration in the economics of technological change’, Econometrica, Vol. 25. Griliches, Z. (1979), ‘Issues in Assessing the Contribution of Research and Development to Productivity Growth’, Bell Journal of Economics 10, 92-116. Grübler, A. (1998), Technology and global change, Cambridge University Press. Hall, P. (1994), Innovation, Economics & Evolution, Harvester Wheatsheaf. Janssen, M. (1998), Modelling global change, Edward Elgar Publishing Limited. Johansen, L. (1960), A Multi-Sectoral Study of Economic Growth. Amsterdam, The Netherlands: North-Holland. Köhler, J. (2003), Long run technical change in an energy-environmenteconomy (E3) model for an IA system: A model of Kondratiev Waves, Integrated Assessment, 4(2), pp. 126-133. Jorgenson, D. (1998), Growth.Vol. 2., chapter 2, The MIT Press. Jorgenson, D. and B. Fraumeni (1981), ‘Relative Prices and Technical Change’, in E. Bernt and B. Field eds., Modeling and Measuring Natural Resource Substitution, MIT Press. Leontief, W. and F. Duchin (1986), The future impact of automation on workers, Oxford University Press. Los, B. (2001), ‘Endogenous growth and structural change in a dynamic Input-output model’, Economic Systems Research, Vol. 13. McFadden, D. (1963), Further Results on CES Production Functions. Review of Economic Studies 30 (2), no. 83 (June): 73-83. Miernyk, W. (1977), ‘The Projection of Technical Coefficients for Medium-Term Forecasting’, in W. Gossling eds., Medium-Term Dynamic Forecasting: the 1975 London Input-output Conference, William Clowes & Sons Limited, London. Mohnen, P. (1997), ‘Introduction: Input-Output Analysis of Interindustry R&D Spillovers’, Economic Systems Research 9, 3-8.

Nelson, R. and S. Winter (1974), ‘Neoclassical vs. Evolutionary Theories of Economic Growth : Critique and Prospectus’, Economic Journal 84 (336), 886-905. Nelson, R. and S. Winter (1982), An Evolutionary Theory of Economic Change, Harvard University Press. Perez C., 1983, Structural change and the Assimilation of New Technologies in the Economic and Social System, Futures, 15 pp.357-75. Romer, P. (1986), ‘Increasing Returns and Long-Run Growth’, Journal of Political Economy 98, S71-S102. Silverberg, G. (1988), ‘Modelling Economic Dynamics and Technical Change: Mathematical Approaches to Self-Organisation and Evolution’, in Dosi, G. et al., eds., Technical Change and Economic Theory, Printer Publishers Ltd., London. Silverberg, G. (1997), ‘Evolutionary Modeling in Economics: Recent History and Immediate Prospects’, for the Workshop on Evolutionary Economics as a Scientific Research Programme, Stockholm. Solow, R. (1956), ‘A Contribution to the Theory of Economic Growth’, Quarterly Journal of Economics 70(1): 65-94. Uzawa, H. (1962), Production Functions with Constant Elasticities of Substitution. Review of Economic Studies 29 (4), no. 81 (October): 291-299. Van den Bergh, J. and J. Gowdy, (2000), ‘Evolutionary Theories in Environmental and Resource Econmics: Approaches and Application’, Environmental and Resource Economics 17: 37-57.
The “evolution” here means the change of economic structure in accordance with technological progress. It may differ from the evolution proposed in typical evolutionary economics, which adopts the Darwinian evolution in biology.


The trans-disciplinary Tyndall Centre for Climate Change Research undertakes integrated research into the long-term consequences of climate change for society and into the development of sustainable responses that governments, business-leaders and decision-makers can evaluate and implement. Achieving these objectives brings together UK climate scientists, social scientists, engineers and economists in a unique collaborative research effort. Research at the Tyndall Centre is organised into four research themes that collectively contribute to all aspects of the climate change issue: Integrating Frameworks; Decarbonising Modern Societies; Adapting to Climate Change; and Sustaining the Coastal Zone. All thematic fields address a clear problem posed to society by climate change, and will generate results to guide the strategic development of climate change mitigation and adaptation policies at local, national and global scales. The Tyndall Centre is named after the 19th century UK scientist John Tyndall, who was the first to prove the Earth’s natural greenhouse effect and suggested that slight changes in atmospheric composition could bring about climate variations. In addition, he was committed to improving the quality of science education and knowledge. The Tyndall Centre is a partnership of the following institutions: University of East Anglia UMIST Southampton Oceanography Centre University of Southampton University of Cambridge Centre for Ecology and Hydrology SPRU – Science and Technology Policy Research (University of Sussex) Institute for Transport Studies (University of Leeds) Complex Systems Management Centre (Cranfield University) Energy Research Unit (CLRC Rutherford Appleton Laboratory) The Centre is core funded by the following organisations: Natural Environmental Research Council (NERC) Economic and Social Research Council (ESRC) Engineering and Physical Sciences Research Council (EPSRC) UK Government Department of Trade and Industry (DTI) For more information, visit the Tyndall Centre Web site ( or contact: External Communications Manager Tyndall Centre for Climate Change Research University of East Anglia, Norwich NR4 7TJ, UK Phone: +44 (0) 1603 59 3906; Fax: +44 (0) 1603 59 3901 Email:

Recent Working Papers Tyndall Working Papers are available online at Mitchell, T. and Hulme, M. (2000). A Countryby-Country Analysis of Past and Future Warming Rates, Tyndall Centre Working Paper 1. Hulme, M. (2001). Integrated Assessment Models, Tyndall Centre Working Paper 2. Berkhout, F, Hertin, J. and Jordan, A. J. (2001). Socio-economic futures in climate change impact assessment: using scenarios as 'learning machines', Tyndall Centre Working Paper 3. Barker, T. and Ekins, P. (2001). How High are the Costs of Kyoto for the US Economy?, Tyndall Centre Working Paper 4. Barnett, J. (2001). The issue of 'Adverse Effects and the Impacts of Response Measures' in the UNFCCC, Tyndall Centre Working Paper 5. Goodess, C.M., Hulme, M. and Osborn, T. (2001). The identification and evaluation of suitable scenario development methods for the estimation of future probabilities of extreme weather events, Tyndall Centre Working Paper 6. Barnett, J. (2001). Security and Climate Change, Tyndall Centre Working Paper 7. Adger, W. N. (2001). Social Capital and Climate Change, Tyndall Centre Working Paper 8. Barnett, J. and Adger, W. N. (2001). Climate Dangers and Atoll Countries, Tyndall Centre Working Paper 9. Gough, C., Taylor, I. and Shackley, S. (2001). Burying Carbon under the Sea: An Initial Exploration of Public Opinions, Tyndall Centre Working Paper 10. Barker, T. (2001). Representing the Integrated Assessment of Climate Change, Adaptation and Mitigation, Tyndall Centre Working Paper 11. Dessai, S., (2001). The climate regime from The Hague to Marrakech: Saving or sinking the Kyoto Protocol?, Tyndall Centre Working Paper 12. Dewick, P., Green K., Miozzo, M., (2002). Technological Change, Industry Structure and the Environment, Tyndall Centre Working Paper 13. Shackley, S. and Gough, C., (2002). The Use of Integrated Assessment: An Institutional Analysis Perspective, Tyndall Centre Working Paper 14. Köhler, J.H., (2002). Long run technical change in an energy-environment-economy (E3) model for an IA system: A model of Kondratiev waves, Tyndall Centre Working Paper 15. Adger, W.N., Huq, S., Brown, K., Conway, D. and Hulme, M. (2002). Adaptation to climate change: Setting the Agenda for Development Policy and Research, Tyndall Centre Working Paper 16. Dutton, G., (2002). Hydrogen Energy Technology, Tyndall Centre Working Paper 17. Watson, J. (2002). The development of large technical systems: implications for hydrogen, Tyndall Centre Working Paper 18. Pridmore, A. and Bristow, A., (2002). The role of hydrogen in powering road transport, Tyndall Centre Working Paper 19. Turnpenny, J. (2002). Reviewing organisational use of scenarios: Case study - evaluating UK energy policy options, Tyndall Centre Working Paper 20. Watson, W. J. (2002). Renewables and CHP Deployment in the UK to 2020, Tyndall Centre Working Paper 21. Watson, W.J., Hertin, J., Randall, T., Gough, C. (2002). Renewable Energy and Combined Heat and Power Resources in the UK, Tyndall Centre Working Paper 22. Paavola, J. and Adger, W.N. (2002). Justice and adaptation to climate change, Tyndall Centre Working Paper 23. Xueguang Wu, Jenkins, N. and Strbac, G. (2002). Impact of Integrating Renewables and CHP into the UK Transmission Network, Tyndall Centre Working Paper 24

Xueguang Wu, Mutale, J., Jenkins, N. and Strbac, G. (2003). An investigation of Network Splitting for Fault Level Reduction, Tyndall Centre Working Paper 25 Brooks, N. and Adger W.N. (2003). Country level risk measures of climate-related natural disasters and implications for adaptation to climate change, Tyndall Centre Working Paper 26 Tompkins, E.L. and Adger, W.N. (2003). Building resilience to climate change through adaptive management of natural resources, Tyndall Centre Working Paper 27 Dessai, S., Adger, W.N., Hulme, M., Köhler, J.H., Turnpenny, J. and Warren, R. (2003). Defining and experiencing dangerous climate change, Tyndall Centre Working Paper 28 Brown, K. and Corbera, E. (2003). A MultiCriteria Assessment Framework for CarbonMitigation Projects: Putting “development” in the centre of decision-making, Tyndall Centre Working Paper 29 Hulme, M. (2003). Abrupt climate change: can society cope?, Tyndall Centre Working Paper 30 Turnpenny, J., Haxeltine A. and O’Riordan, T. (2003). A scoping study of UK user needs for managing climate futures. Part 1 of the pilot-phase interactive integrated assessment process (Aurion Project), Tyndall Centre Working Paper 31 Xueguang Wu, Jenkins, N. and Strbac, G. (2003). Integrating Renewables and CHP into the UK Electricity System: Investigation of the impact of network faults on the stability of large offshore wind farms, Tyndall Centre Working Paper 32 Pridmore, A., Bristow, A.L., May, A. D. and Tight, M.R. (2003). Climate Change, Impacts, Future Scenarios and the Role of Transport, Tyndall Centre Working Paper 33 Dessai, S., Hulme, M (2003). Does climate policy need probabilities?, Tyndall Centre Working Paper 34 Tompkins, E. L. and Hurlston, L. (2003). Report to the Cayman Islands’ Government. Adaptation lessons learned from responding to tropical cyclones by the Cayman Islands’ Government, 1988 – 2002, Tyndall Centre Working Paper 35

Kröger, K. Fergusson, M. and Skinner, I. (2003). Critical Issues in Decarbonising Transport: The Role of Technologies, Tyndall Centre Working Paper 36 Ingham, A. and Ulph, A. (2003) Uncertainty, Irreversibility, Precaution and the Social Cost of Carbon, Tyndall Centre Working Paper 37 Brooks, N. (2003). Vulnerability, risk and adaptation: a conceptual framework, Tyndall Centre Working Paper 38 Tompkins, E.L. and Adger, W.N. (2003). Defining response capacity to enhance climate change policy, Tyndall Centre Working Paper 39 Klein, R.J.T., Lisa Schipper, E. and Dessai, S. (2003), Integrating mitigation and adaptation into climate and development policy: three research questions, Tyndall Centre Working Paper 40 Watson, J. (2003), UK Electricity Scenarios for 2050, Tyndall Centre Working Paper 41 Kim, J. A. (2003), Sustainable Development and the CDM: A South African Case Study, Tyndall Centre Working Paper 42 Anderson, D. and Winne, S. (2003), Innovation and Threshold Effects in Technology Responses to Climate Change, Tyndall Centre Working Paper 43 Shackley, S., McLachlan, C. and Gough, C. (2004) The Public Perceptions of Carbon Capture and Storage, Tyndall Centre Working Paper 44 Purdy, R. and Macrory, R. (2004) Geological carbon sequestration: critical legal issues, Tyndall Centre Working Paper 45 Watson, J., Tetteh, A., Dutton, G., Bristow, A., Kelly, C., Page, M. and Pridmore, A., (2004) UK Hydrogen Futures to 2050, Tyndall Centre Working Paper 46 Berkhout, F., Hertin, J. and Gann, D. M., (2004) Learning to adapt: Organisational adaptation to climate change impacts, Tyndall Centre Working Paper 47 Pan, H. (2004) The evolution of economic structure under technological development, Tyndall Centre Working Paper 48