Development and climate change (and implications for the modelling of adaptation)

Transcription

1 Development and climate change (and implications for the modelling of adaptation) Franck LECOCQ 1,2 1 AgroParisTech, Engref UMR 356 Economie Forestière, F54000 Nancy, France 2 INRA, UMR 356 Economie Forestière, F Nancy, France Expert workshop on the economics of adaptation Session 4B Modelling adaptation Economy wide effects OECD, Paris, 7-8 April 2008

2 Introduction Objective: Draw implications of ongoing research on the relationships between climate change and development for the modelling of adaptation Source: one paper on the implications of climate change for economic growth, and one paper on the balance between mitigation and adaptation Both papers are jointly written with Dr. Zmarak Shalizi (Independent Scholar, formerly World Bank DECRG)

3 The global decision problem: Minimize the climate change bill Four components: mitigation, proactive adaptation, reactive adaptation and remaining ultimate damages The distinction between proactive and reactive adaptation is important for political economy reasons The benefits of adaptation can only be measured in terms of avoided damages, for lack of other overall performance yardstick; hence the importance of representing impacts The decision problem is complicated at the national level because mitigation (and some elements of adaptation) is a global public goods. We do not address the resulting cooperation issues.

4 Mitigation, reactive adaptation and proactive adaptation are not independent Mitigation and proactive adaptation are usually imperfect substitutes I.e., the marginal benefits of proactive adaptation depends on the level of mitigation, and vice versa This is independent from, and in addition to, joint production of mitigation and adaptation Mitigation and proactive adaptation are imperfect substitutes because only mitigation can avoid the irreversible impacts of climate change Similarly, reactive adaptation and proactive adaptation are sometimes substitutes, sometimes complements Hence the importance of integrated modeling with all four components of the climate change bill

5 The proper counterfactual is no action and full damages

6 The different components of the climate change bill impact economic growth via different channels Nature Timing Additional expenditures Direct impact on prod. factors (and their productivity) Might be smoothed out over time Mitigation, proactive adaptation Tend to be concentrated in time Reactive adaptation Remaining ultimate damages

7 Implications of the climate change bill for economic growth in neoclassical growth models Neoclassical (i.e. Cass-Koopmans) growth models simplest analytical framework to discuss impacts on economic growth, also underlie most numerical models on climate change Variations in stocks / productivity of production factors affect transitional, but generally not long-run economic growth rates The structure of the steady-state growth path, however, may vary (notably the capital to labor and consumption to labor ratios) Variations in expenditures have typically only transitional impacts, except when proportional to output In the latter case, it is as if the marginal productivity of the whole economy was affected Three key reasons for the limited effects of climate change on economic growth in the neoclassical model: Each factor has decreasing returns when others held constant The engines of growth (demographics and technological change) are exogenous All markets clear instantaneously

8 The Cass-Koopmans has limited explanatory power of observed growth patterns Empirical studies consistently suggest that factor accumulation explains no more than half of observed growth rates, particularly in developing countries (e.g., Easterly and Levine, 2001, O Connell and Ndulu, 2000) There is no theoretical or empirical consensus on the determinants of the residual (Total Factor Productivity) Four sets of factors play a role: Structural factors (e.g., changing shares of agriculture and industry) Geographical factors (e.g., landlocked vs. non-landlocked) Shocks, of whatever nature (e.g., terms of trade, financial crises, natural hazards) Policy and institutional factors (e.g., macroeconomic policy, etc.)

9 Climate change in multi-sector growth models: The importance of spillovers and factors mobility In neoclassical growth models with two or more sectors, impacts on one sector propagate to the other. Under decreasing returns and full factor mobility assumptions, impacts in one sector cannot be fully compensated by positive implications in another By design, multi-sector models tend to be numerical, which raises the question of the extent to which their results depend on particular modeling assumptions, or on the very structure of the growth engine Yet understanding the reasons that underpin particular numerical results is very important for policy purposes

10 The interaction between shocks and rigidities leads to higher transitional costs Observed costs of shocks for growth are rather high in many economies and play a significant role in explaining cross-country economic performances (e.g., Easterly et al., 1993) IMF (2003): large-scale natural disasters have occurred on average once every 20 years in LDCs over the period, with an average impact on GDP ranging from -2.7% to -5.8% in the period Can be explained by dependency on climate-sensitive sectors, rigidities (e.g., in price adjustment, factor mobility or in absorptive capacity e.g., Hallegatte et al. 2007), and institutional factors

11 Taking (local) non-convexities into account has important but ambiguous implications for the cc/growth relationship Poverty trap models suggest that shocks may have longlasting consequences by reducing the chance that a country grows out of the trap (or increasing the chance that it falls into it) In endogenous growth models, climate change impacts on population have potentially larger consequences for growth, at least in the short- and medium-term, but not necessarily in the long-term. A consensus seems to be emerging that induced technological change is an important driver of costs of mitigation (and probably adaptation) policies, especially in the long-run

12 Taking (local) non-convexities into account has important but ambiguous implications for the cc/growth relationship Increasing returns also create the possibility of lock-ins, with potentially major but ambiguous implications for the climate change / growth relationship Growth benefits from geographical concentration. Thus, localized impacts of climate change on a growth nucleus could have disproportionate impacts on economic growth Similarly, implications for growth of impacts on infrastructure that links growth engines to markets (roads, ports) may be very large

13 Conclusion (1) Some insights for modelling adaptation The proper counterfactual is a world with full climate change The optimal levels of mitigation, proactive adaptation, reactive adaptation and ultimate damages cannot be discussed independently The implications of proactive adaptation and reactive adaptation for growth will strongly depend on the underlying growth model The Cass-Koopmans neoclassical framework does not incorporate real-world growth processes Though more research is required on key elements of growth (notably ridigities and lock-ins), there is already some material in the growth literature that numerical models can rely on beyond the CK framework

14 Conclusion (2): Some important policy questions left out How adaptation differs from regular development aid? The distinction is easy to make in theory, and might be made in some models, but it is difficult in practice To which extent should the public sector finance adaptation? Unless the private and public benefits of adaptation are somehow distinguished, models may not be able to address this question How does one go around budget constraints? Models tend to assume complete transferability of resources across regions/sectors but this is not always the case In addition, climate shocks might reduce the fiscal base, thereby increasing risks of budget constraints at the time resources are needed for reactive adaptation

15 Thank you very much!