Can the Paris Agreement Support Achieving the Sustainable Development Goals?

Abstract

This chapter provides an ex-ante, quantitative assessment of the synergies and trade-offs between the implementation of the Paris Agreement and sustainable development. It develops a framework for comparing historical and future sustainability performance that combines a Computable General Equilibrium model for describing future global and regional baseline and policy scenarios to 2030 with empirically-estimated relationships between macroeconomic variables and sustainability indicators. Results indicate that the commitments submitted within the Paris Agreement reduce the gap toward a sustainable 2030 in all regions, but heterogeneity across regions and sustainability indicators call for complementary sustainable development polices.

Appendices

Appendix 1

The current list of SDG indicators defined by the works of UN Inter-agency Expert Group on SDG Indicators (United Nations (UN) 2017a) considers 232 indicators. The well-established and in use indicators are less than half of them. The final ASDI screening considers 27 indicators covering 16 SDGs (All but SDG5—Achieve gender equality and empower all women and girls). Table 2 lists the ASDI indicators coupled with the related SDG, the sustainability pillar of pertinence, whether they derive from a regression (regression results are reported in Table 3), how they are computed and the main source of data.

Table 2 ASDI indicators

Table 3 Regression table

In order to compare country performance in different SDG indicators and to compute some aggregate measures, it is necessary to bring all indicators to a common measurement unit, the [0,100] scale (normalisation). The normalisation is obtained using a benchmarking procedure that defines two threshold values for each indicator: unsustainable and sustainable levels. In choosing the threshold levels, we firstly looked at the 169 SDG targets, which are our preferred source if they provide quantitative targets. When the targets are qualitative, other sources are preferred such as policy targets in OECD or countries’ best practices.

Table 4 shows the threshold values used for the normalisation process.

Table 4 ASDI benchmarks

ASDI framework considers several aggregation steps in order to produce aggregate indices conveying more synthetic information to policymakers:

• SDG indices are the average value of indicators pertaining to each goal;

• The ASDI index is the average of scores among all SDGs.

Appendix 2

Model Description

ICES is a recursive-dynamic multiregional Computable General Equilibrium (CGE) model developed to assess the impacts of climate change on the economic system and to study mitigation and adaptation policies (Eboli et al. 2010). The model’s general equilibrium structure allows for the analysis of market flows within a single economy and international flows with the rest of the world. This implies going beyond the simple quantification of direct costs, to offer an economic evaluation of second and higher-order effects within specific scenarios either of climate change, climate policies or different trade and public-policy reforms in the vein of conventional CGE theory. The core structure of ICES derives from the GTAP-E model (Burniaux and Truong 2002), which in turn is an extension of the standard GTAP model (Corong et al. 2017). The General Equilibrium framework makes it possible to characterise economic interactions of agents and markets within each country (production and consumption) and across countries (international trade). Within each country the economy is characterised by a number of industries n, a representative household and the government. Industries are modelled as representative cost-minimising firms, taking input prices as given. In turn, output prices are given by average production costs. The production functions (Fig. 5) are specified via a series of nested Constant Elasticity of Substitution (CES) functions. In the first nest, a Value-Added-Energy nest (QV AEN) (primary factors, i.e. natural resources, land, and labour and a Capital+Energy composite), is combined with intermediates (QF), in order to generate the output. Perfect complementarity is assumed between value added and intermediates. This implies the adoption of a Leontief production function. For sector i in region r final supply (output) results from the following constrained production cost minimization problem for the producer:

$$\displaystyle \begin{aligned} min \quad PVAEN_{i,r} * QVAEN_{i,r} + PF_{i,r} * QF_{i,r} \\ s.t. \quad Y_{i,r}= min (QVAEN_{i,r} , QF_{i,r} )\end{aligned} $$

where PV AEN and PF are prices of the related production factors.

Fig. 5

ICES production tree

The second nested-level in Fig. 5 (left hand side of the production tree) includes the value added and the energy composite (QV AEN). This composite stems from a CES function that combines four primary factors: land (QLAND), natural resources (QFE), labour (QFE) and the capital-energy bundle (QKE) using σ_VAE as elasticity of substitution. Primary factor demand on its turn derives from the first order conditions of the following constrained cost minimization problem for the representative firm:

$$\displaystyle \begin{aligned} min \quad P_{i,r}^{Land}* LAND_{1,r} + P_{i,r}^{NR} *NR_{i,r} + P_{i,r}^{L}*L_{i,r} + P_{i,r}^{KE} *KE_{i,r} \\ s.t. \quad QVAEN_{i,r}= (LAND_{i,r}^{\frac{\sigma_{VAE}-1}{\sigma_{VAE}}} + NR_{i,r}^{\frac{\sigma_{VAE}-1}{\sigma_{VAE}}} + L_{i,r}^{\frac{\sigma_{VAE}-1}{\sigma_{VAE}}} + KE_{i,r}^{\frac{\sigma_{VAE}-1}{\sigma_{VAE}}}) ^{\frac{\sigma_{VAE}-1}{\sigma_{VAE}}} \end{aligned} $$

On its turn, the KE bundle combines capital with a set of different energy inputs. This is peculiar to GTAP-E and ICES. In fact, energy inputs are not part of the intermediates, but are associated to capital in a specific composite. The energy bundle is modelled as an aggregate of electric and non-electric energy carriers. Electricity sector differentiates between intermittent and non-intermittent sources. Wind and solar, which are intermittent sources, are separated from non-intermittent sources: hydro power and the rest of electricity produced using fossil fuel sources (coal, oil, and gas).¹² The Non-Electric bundle is a composite of nuclear and non-nuclear energy. The aggregate Non-nuclear energy combines, in a series of subsequent nests, Coal, Natural Gas, Crude Oil, Petroleum Products, and Biofuels, while Nuclear corresponds to the carrier used for electricity generation. All elasticities regarding the inter-fuel substitution bundles are those from GTAP-E (Burniaux and Truong 2002), while for the extended renewable electricity sectors we set those values considering different studies (Paltsev et al. 2005; Bosetti et al. 2006). The demand of production factors (as well as that of consumption goods) can be met by either domestic or foreign commodities which are, however, not perfectly substitute according to the “Armington” assumption. In general, inputs grouped together are more easily substitutable among themselves than with other elements outside the nest. For example, the substitutability across imported goods is higher than that between imported and domestic goods. Analogously, composite energy inputs are more substitutable with capital than with other factors. In ICES, two industries are treated in a special way and are not related to any country, viz. international transport and international investment production. International transport is a world industry, which produces the transportation services associated with the movement of goods between origin and destination regions, thereby determining the cost margin between f.o.b. and c.i.f. prices. Transport services are produced by means of factors submitted by all countries, in variable proportions. In a similar way, a hypothetical world bank collects savings from all regions and allocates investments in order to achieve equality in the absolute change of current rates of return.

Fig. 6

Sources and uses of regional household income

Figure 6 describes the main sources and uses of regional income. In each region, a representative utility maximising household receives income, originated by the service value of national primary factors (natural resources, land, labour, and capital) that it owns and sells to the firms. Capital and labour are perfectly mobile domestically but immobile internationally (investment is instead internationally mobile). Land and natural resources, on the other hand, are industry-specific. The regional income is used to finance aggregate household consumption and savings.

Government income equals to the total tax revenues from both private household and productive sectors, a series of international transactions among governments (foreign aid and grants) and national transfers between the government and the private (Delpiazzo et al. 2017). Both the government and the private household consume and save a fraction of their income according to a Cobb-Douglas function. The government income not spent is saved, and the sum of public and private savings determines the regional disposable saving, which enters the Global Bank as in the core ICES. Both private and public sector consumption are addressed to all commodities produced by each firm/sector. Public consumption is split into a series of alternative consumption commodities according to a Cobb-Douglas specification. However, almost all public expenditure is concentrated in the specific sector of Non-market Services, including education, defence, and health. Private consumption is analogously addressed towards alternative goods and services including energy commodities that can be produced domestically or imported. The functional specification used at this level is the Constant Difference in Elasticities (CDE) form: a non-homothetic function, which is used to account for possible differences in income elasticities for the various consumption goods.¹³

Fig. 7

Recursive-dynamic feature of ICES model

The recursive-dynamic feature is described in Fig. 7. Starting from the picture of the world economy in the benchmark year, following socioeconomic (e.g. population, primary factors stocks, and productivity) as well as policy-driven changes occurring in the economic system, agents adjust their decisions in terms of input mix (firms), consumption basket (households), and savings. The model finds a new general (worldwide and economy-wide) equilibrium in each period, while all periods are interconnected by the accumulation process of physical capital stock, net of its depreciation. Capital growth is standard along exogenous growth theory models and follows:

$$\displaystyle \begin{aligned} Ke_{r} = I_{r} + (1-\delta) Kb_{r} \end{aligned} $$

where Ke_r is the end of period capital stock, Kb_r is the beginning of period capital stock, δ is capital depreciation and I_r is endogenous investment. Once the model is solved at a given step t, the value of Ke_r is stored in an external file and used as the beginning of period capital stock of the subsequent step t+1. The matching between savings and investments only holds at the world level; a fictitious world bank collects savings from all regions and allocates investments following the rule of highest capital returns.

As with capital, at each simulation step the government net deficit at the end of the period is stored in an external file and adds up to next year debt.

Regional Aggregation

ICES is a Computable model: all the model behavioural equations are connected to the GTAP 8 database (Narayanan and McDougall 2012), which collects national social accounting matrices from all over the world and provides a snapshot of all economic flows in the benchmark year. Being based on the GTAP database, ICES has worldwide coverage. In this analysis, we consider 45 countries/regions (Fig. 8).

Fig. 8

Regional aggregation ICES model

For sake of clarity in presenting results, we further aggregate the 45 countries/regions in eight regional aggregates following the mapping presented in Table 5.

Table 5 Mapping ICES regions into macro regional aggregates

Reference Scenario

Our reference in designing the baseline scenario is the set of possible futures envisioned by the climate change community and known as Shared Socioeconomic Pathways (SSPs) (O’Neill et al. 2017). These are five possible futures with different mitigation/adaptation challenges and are characterised by different evolution of main socioeconomic variables. SSPs can be linked to Representative Concentration Pathways (RCPs), that envisions the GHG emission evolution and forcing and temperature rise due to specific patter of socioeconomic growth (Riahi et al. 2017). SSPs provide future patterns for population, working age population and GDP at country level. Other trends for exogenous drivers such as primary factor productivity, sector-specific efficiency, total factor productivity, and energy prices are then used in order to calibrate given endogenous variables, namely GDP, energy use, emissions and value-added shares to be coherent to the selected RCP.

Among Shared Socioeconomic Pathways (SSPs), we used as business as usual SSP2 “Middle of the Road” scenario. The main features of this scenario are:

• similar trends of recent decades, but some progresses towards achieving development goals;

• medium population growth;

• per capita income levels grow globally at a medium pace; slow income convergence across countries; some improvements in the intra-regional income distributions;

• reductions in resource and energy intensity, and slowly decreasing fossil fuel dependency.

Give the short time horizon of the proposed analysis, we focus on the SSP2 because it is the the pathway that more closely follows the historical development in terms of socioeconomic variables evolution (medium population and GDP growth). In calibrating the SSP2, we followed not only the socioeconomic trends reported in SSP database,¹⁴ but we also adjusted energy efficiency and fuel prices in order to obtain a global emission level in line with IAM multi-model projections (Riahi et al. 2017). The literature reports a range between 61,279 and 70,005 Mt CO₂-eq/year in 2030. Our baseline global emissions are 65,140 Mt CO₂-eq/year. The projected emission range in 2100 is between 85,030 and 106,778 Mt CO₂-eq/year which corresponds to a radiative forcing between 6.561 and 7.251 W/m², and a temperature rise between 3.8 and 4.2 ^∘C. These results place our baseline in between RCP6 and RCP8.5.

Mitigation Scenario

We designed a mitigation scenario mimicking Paris Agreement functioning: all parties achieve the conditional mitigation targets stated in the NDC by 2030; for regional aggregates, we computed reference and target emission levels and calculated the required regional reduction. We relies on CAIT database for computing reference historical emission levels, whereas our baseline scenario is used when NDC uses a BAU scenario as term of comparison. Due to model limitations, we impose the GHG emission targets only to CO₂ emissions. Mitigation objectives considered for each country/region are reported in Table 6. Two countries in our aggregation do not have a clear quantitative mitigation target, i.e. Egypt and Bolivia; therefore, in our simulation, we assume they have not a NDC.

Table 6 Emission reduction target in 2030

The mitigation policy starts in 2013 and it is fully achieved by 2030. The European Union (EU28) opts for an Emission Trading System (ETS), while all other countries achieve their contributions unilaterally with a domestic carbon tax. China, India, and Chile have expressed their NDCs in terms of emission intensity. Carbon tax revenues are redistributed internally to government investment, public saving and transfers to households.