1 The Importance of Wooden Biomass in the Transition to a Bioeconomy in Latvia 1 Vineta Tetere 1,2 *, Jack Peerlings 1, Liesbeth Dries 1 2 * Corresponding author, vineta.tetere@wur.nl 3 1 Wageningen University, the Netherlands 4 2 Latvia University of Life Sciences and Technologies, Latvia 5 This article has been accepted for publication and undergone full peer review but has not been 6 through the copyediting, typesetting, pagination and proofreading process, which may lead to 7 differences between this version and the Version of Record. 8 Please cite this article as: 9 Tetere V., Peerlings J., Dries L. (2025). The Importance of Wooden Biomass in the Transition 10 to a Bioeconomy in Latvia, Bio-Based and Applied Economics, Just Accepted. 11 DOI:10.36253/bae-16107 12 Abstract 13 The EU Green Deal advocates decarbonising the EU’s energy sector, largely by transitioning 14 to renewable sources. Latvia aims to increase the share of renewable energy production in total 15 energy production to 50% by 2030 (it was 39% in 2017), prioritising biomass from forests and 16 wood for bioenergy. This paper evaluates increasing the tax on non-biobased energy use 17 alongside implementing a subsidy for biobased energy use, particularly from wood biomass, to 18 promote the substitution of the first by the latter as a step towards climate neutrality and energy 19 self-sufficiency. Furthermore, it examines technological advancements in the bioenergy sector 20 as an alternative instrument. Using an applied general equilibrium model and 2015 supply and 21 use data, the study allows for substitution between domestic and imported inputs and between 22 the non-biobased and biobased energy product. Given Latvia’s heavy reliance on imported 23 fossil fuels, these measures could lead to a 58% increase in bioenergy production compared to 24 2015, reducing CO2 emissions by 0.3 – 1.7%, and reducing non-biobased energy imports by 25 2.5-4.2%. 26 Keywords: biobased energy, non-biobased energy, product tax, subsidy 27 JEL codes: C68, E17, Q23 28 mailto:vineta.tetere@wur.nl 2 1. Introduction 29 The transition from a fossil-based economy to a biobased economy is considered a priority to 30 mitigate the effects of climate change in the European Union (EU). The EU Green Deal states 31 that the EU has to become climate neutral by 2050 (EC b, n.d.). This requires the EU to reduce 32 net greenhouse gas (GHG) emissions to zero. Carbon dioxide (CO2) remains the main 33 greenhouse gas emitted through human activities, and most CO2 emissions come from the 34 energy sector: electricity, heating, and transport. Therefore, one of the main actions proposed 35 in the Green Deal is to decarbonise the EU’s energy sector, largely through the transition of the 36 generation of power from fossil-based to renewable sources. The Latvian government follows 37 this action with the intention to increase the production of energy from renewable sources to 38 50% of the total energy production in 2030 (it was 39% in 2017). To achieve this goal, an 39 emphasis will be placed on sourcing biomass from the forest and wood industry to be used for 40 the production of biobased energy (EM, 2019). The reason for this focus is that the production 41 and use of other renewable energy products are small. For example, in 2015 91.7% of the use 42 of renewable energy was from biomass from forests and wood, for hydropower, this was 7.7% 43 and for wind 0.6%. 44 As a reaction to the Russian invasion of Ukraine, the European Commission introduced the 45 REPowerEU plan in March 2022 outlining measures to drastically reduce Russian gas imports 46 and achieve independence from Russian fossil fuels before the end of the decade. The key 47 elements in this plan are diversifying supplies, reducing demand, and increasing the production 48 of green energy in the EU. This is expected to accelerate the green transition by reducing GHG 49 emissions, reducing dependency on imported fossil fuels, and protecting the EU against price 50 hikes on the energy market (EU Commission, 2022). 51 Economic instruments like taxes on fossil fuels and subsidies on biobased energy production 52 or use can contribute to achieving the goals of the Green Deal and the REPowerEU plan. Since 53 fossil fuels are non-renewable and GHG emissions are harmful to the environment, a product 54 3 tax can help in reducing its demand and supply, as it increases the net price demanders pay and 55 decreases the price suppliers receive. Product subsidies for biobased energy have the opposite 56 effect. Moreover, product taxes and subsidies can stimulate the development and use of more 57 sustainable technologies (Wolfson & Koopmans, 1996). 58 A tax on fossil energy can - in addition to the reduction in emissions – lead to an increase in 59 welfare if it reduces the tax distortions caused by other taxes (“second best effect”). If this 60 happens, then we speak of a ‘double dividend’ (see De Mooij, 2002 and Goulder, 1995). A 61 potential double dividend can be an extra incentive to introduce a tax on fossil energy use. 62 In Latvia, the forest sector is one of the cornerstones of the economy. Forestry, wood processing 63 and furniture manufacturing contributed 5.1% to GDP, 5.4% to total employment and 20.7% to 64 exports in 2018 (AM - Ministry of Agriculture, 2022). Furthermore, the forest area covers 52% 65 of the country’s territory and this is expanding. It has doubled since 1935 due to farm 66 abandonment that resulted in the conversion of cropland fields into young forests (Fonji & Taff, 67 2014). The increase in forest area is expected to continue because of purposeful afforestation, 68 as well as through the continued natural overgrowth of forest on non-agricultural lands. 69 Moreover, wooden biomass is increasing annually due to more sustainable forest management 70 in recent decades (Lazdiņš et al., 2019). This opens possibilities for further increase in the use 71 of wooden biomass in the production of biobased energy. 72 The aim of this paper is to investigate the potential of taxes, subsidies, and technological change 73 to increase the share of biobased energy production in Latvia. More specifically, it assesses the 74 effect of an increase in the tax on non-biobased energy use and the implementation of a subsidy 75 on biobased energy use, especially from wooden biomass, to facilitate the substitution of the 76 first for the latter as a step in the transition of Latvia’s economy towards climate neutrality and 77 self-sufficiency of energy. Moreover, the paper assesses the effects of technological change in 78 the industry producing biobased energy. Such a technological change in the production of 79 biobased energy is instrumental for a successful transition of the energy sector. To this end, the 80 4 EU and the Latvian government will stimulate technological change as part of the Green Deal 81 using € 4.4 billion from EU funds for Latvia between 2021 and 2027 (FM, n.d.). 82 The paper uses an applied general equilibrium (AGE) model based on the model developed by 83 Komen and Peerlings (1999) to achieve the aim. Their model included greenhouse gas 84 emissions and policy scenarios to reduce these emissions. However, it did not distinguish 85 between non-biobased and biobased energy, nor did it include biomass. Especially in the 1980s 86 and 1990s AGE models were used to assess different policy issues, e.g., agricultural policy 87 reform, environmental taxation, etc. (for an overview, see Bergman, 1990; Gunning & Keyzer, 88 1995; Robinson, 1989). Policy issues simulated with AGE models reflect relatively large 89 shocks to an economy, as AGE models explicitly model the economy as a whole. Calculating 90 the effects of large shocks cannot be done using a partial equilibrium model given that these 91 models assume too many variables (e.g. wages, interest, etc.) exogenous. The transition towards 92 climate neutrality and energy self-sufficiency can be considered as a large shock to the Latvian 93 economy. Data come from the supply and use tables for 2015 and national accounting data from 94 the Latvian Central Statistics Bureau (CSB). It is assumed that a nested production structure 95 allows for explicit imperfect substitution between domestic and imported inputs in energy 96 production to account for Latvia’s current dependence on imported fossil fuels. By increasing 97 the use of (domestically produced) wooden biomass in the energy sector through a tax on the 98 use of non-biobased products and a subsidy on the use of the biobased products, the amount of 99 CO2 emissions from fossil energy and the dependence on fossil energy imports are expected to 100 reduce. The technological change is expected to lead to similar effects. To the best of our 101 knowledge, this is the first application of an AGE model to Latvia and the first AGE analysis 102 to investigate the effects of the Green Deal. 103 The remainder of the paper is structured as follows. Section 2 describes the energy and forestry 104 sectors and policies of Latvia. Section 3 presents the AGE model. The data are described in 105 5 Section 4. Section 5 presents and discusses the results of the model. Section 6 concludes and 106 provides a general discussion. 107 108 2. Energy and forestry sectors, and policies 109 Latvia is highly dependent on imports of fossil fuels. Table 1 shows that oil products and natural 110 gas are not produced in the country. Electricity is produced mostly domestically, partly from 111 fossil fuels and partly from renewable energy sources. 112 Table 1. Energy production, imports, exports and domestic consumption in Latvia, 2020 113 Product Production Imports Exports Domestic consumption Oil products, thousand Euro1 - 726.56 145.64 542.4 Natural gas, million Euro - 433.74 0.00 433.74 Electricity, million Euro 181.86 137.71 84.02 235.55 1: Production plus imports does not add up to domestic consumption and exports because of changes in stocks and 114 statistical issues. 115 Source: CSB, 2021a, 2021b, 2021c 116 To meet the objectives set by the EU in the Green Deal and international commitments (see 117 Table 2), the Latvian government drafted the National Energy and Climate Plan 2021 – 2030. 118 The long-term goal of the plan is to promote the development of a climate-neutral economy in 119 a sustainable, competitive, cost-effective, secure, and market-based way by improving energy 120 security and public welfare. To achieve this goal, it is necessary: ‘1) To promote the efficient 121 use of resources, as well as their self-sufficiency and diversity; 2) To ensure a significant 122 reduction in the consumption of resources, in particular fossil and unsustainable resources, 123 6 and a simultaneous transition to sustainable, renewable and innovative use of resources, 124 ensuring equal access to energy for all sections of society; 3) To stimulate research and 125 innovation that contributes to the development of a sustainable energy sector and the mitigation 126 of climate change’ (EM - Ministry of Economics, 2019). 127 Table 2. EU and Latvia`s energy policy indicators and targets 128 Indicator/target EU’s target, 2030 Latvia’s actual value in 2017 Latvia’s target, 2030 Latvia’s target, 2050 Reducing GHG emissions (% to 1990) (LULUCF* excluded) -40 -57 -65 Climate neutrality (irreducible GHG emissions are compensated by LULUCF sector) Reducing GHG emissions (% to 1990) (LULUCF included) - - -38 Energy produced from RES**, share of gross final consumption (%) 32 39 50 - Share of imports in gross domestic energy consumption (%) - 44.1 30-40 - * Land Use, Land Use Change and Forestry 129 ** Renewable energy sources 130 Source: EM, 2019 131 One of the goals of the National Energy and Climate Plan is to increase the share of renewable 132 energy sources in Latvia. The plan includes the so-called ‘tax greening’ (“polluter pays 133 principle”), where the focus is on taxes such as excise, value added, vehicle, electricity, and 134 natural resource taxes. However, to our knowledge, these have not been implemented by the 135 beginning of 2025. 136 7 The transition from a fossil-based economy to a biobased economy is especially relevant for 137 the forest sector in Latvia. The forest sector is expected to contribute to this transition through 138 the replacement of fossil fuels and non-renewable products with forestry-based products 139 (Kröger & Raitio, 2017). In addition to being used in the production of traditional wood-based 140 products, such as furniture, wooden biomass is increasingly being used in energy generation 141 and in the production of textiles, bioplastics, chemicals, and intelligent packaging, and is also 142 contributing to the construction sector (Hetemäki et al., 2017; Hurmekoski et al., 2018). 143 One fifth of the forest stands in Latvia is in the age of mature and old-growth (CSB, 2021d). 144 The CO2 sequestration capacity of old trees is relatively low, hindering the fulfilment of the 145 Green Deal targets making them a potential feedstock to produce biobased energy. 146 According to data from the EU Bioeconomy Monitoring System Dashboard (EC, n.d.), 58.5% 147 of wooden biomass in Latvia is used in the production of bioenergy and 41.5% is used as 148 materials in manufacturing in 2015. The largest share of wooden biomass in energy production 149 was taken by firewood (30% of the total consumption of energy sources) in 2018, followed by 150 briquettes, pellets, wood scraps, and wood chips. The largest consumers of wooden biomass are 151 households followed by the energy transformation sector (Figure 1). 152 153 Figure 1. Wooden Biomass Consumption in Energy Production in Latvia 2016 – 2019 (%) 154 8 155 Source: AM, 2022 156 157 3. Model 158 This section describes the AGE model developed and applied in this paper. The model is based 159 on the model of Komen and Peerlings (1999). Their model included greenhouse gas emissions 160 and policy scenarios to reduce these emissions. However, it did not distinguish between non-161 biobased and biobased energy. We present the model in Section 3.1 and discuss the modelling 162 of taxes, subsidies, and technological change in section 3.2. A full description can be found in 163 Appendix A. 164 165 3.1 General description 166 An AGE model describes an economy as a whole and is therefore useful to analyse large shocks 167 to the economy that affect, through market linkages, all the economy’s actors (i.e. industries, 168 households, government). It mainly consists of demand and supply functions of commodities 169 27 25 27 26 37 35 35 33 30 33 32 35 7 7 5 6 0 20 40 60 80 100 120 2016 2017 2018 2019 % Year Industry and construction Households Energy transformation sector Other 9 and factor inputs, and income formation and distribution equations (Dervis et al., 1982). The 170 developed model contains 60 commodities and 60 industries including both a non-biobased and 171 biobased energy commodity and industry. However, one industry can produce more than one 172 commodity, and one commodity can be produced by more than one industry. Industries are 173 assumed to minimise costs as they face a constant returns to scale nested Constant Elasticity of 174 Substitution (CES) production function (see, e.g. Arrow et al., 1961; Sato, 1967). 175 Figure 2 shows that for industry b, the intermediate energy intermediate inputs (𝐼𝑁𝑏,𝑔 ∀𝑔 ∈176 𝑆𝑒𝑛), material intermediate inputs (𝐼𝑁𝑏,𝑔 ∀𝑔 ∈ 𝑆𝑚𝑎𝑡) and primary inputs (𝑃𝑅1, 𝑃𝑅2) are 177 aggregated into 3 aggregate inputs respectively. This is done using 3 CES functions, each with 178 their own substitution elasticity. The aggregate inputs are then aggregated into an aggregate 179 output (𝑌𝑏) using a CES function with again its own substitution elasticity. The aggregated 180 output is then divided into different outputs (𝑌𝑌𝑏,𝑔) using a Leontief transformation function 181 (i.e., using fixed ratios). 182 183 Figure 2. Production of industry b 184 185 Where: 186 Inputs Aggregated inputs Aggregated output Outputs YYb,g ∀g ∈ G Yb (CES, Leontief) AENb (CES) INb,g ∀g ∈ Sen AINb (CES) INb,g ∀g ∈ Smat APRb (CES) PR1,PR2 10 INb,g ∀g ∈ Sen: use of energy commodity g as an intermediate input in industry b. Commodities 187 are in the set Sen of energy intermediate inputs. 188 INb,g ∀g ∈ Smat: use of commodity g as an intermediate input in industry b. Commodities are in 189 the set Smat of non-energy intermediate inputs. 190 PR1,PR2: labour (j=1) and capital (j=2) used in industry b. 191 AENb, AINb, and APRb: aggregate energy, aggregate intermediate and aggregate primary input 192 use, respectively, in industry b. 193 𝑌𝑏: aggregate output in industry b. 194 𝑌𝑌𝑏,𝑔: output g of industry b. 195 196 Source: Authors` elaboration 197 198 At the highest level, outputs are produced by an aggregate energy input, an aggregate 199 intermediate input, and an aggregate factor input. At the lowest level, the aggregate energy 200 input is composed of a biobased and a non-biobased energy input. The aggregate intermediate 201 input is composed of 58 non-energy intermediate inputs. The aggregate factor input is 202 composed of labour and capital. Cost minimisation leads to the demand for energy intermediate 203 inputs, non-energy intermediate inputs, labour and capital. 204 Figure 3 shows that in the next step of the model, the outputs produced by different industries 205 are aggregated commodity by commodity. Aggregation gives domestic production (𝐷𝑃𝑔) of a 206 commodity g. Domestic production competes with imports of the same commodity (𝐼𝑀𝑔). This 207 competition can be seen as an aggregation into total supply (𝑆𝑃𝑔) using a CES production 208 function. The total supply is then disaggregated using a CET transformation function into 209 domestic use (𝐷𝑈𝑔) and exports (𝐸𝑋𝑔). CES production and CET transformation functions 210 imply that with profit maximisation relative prices determine demand and supply, respectively. 211 11 Domestic use equals the sum of intermediate demand (∑ 𝐼𝑁𝑏,𝑔𝑏∈𝐵 ), private household demand 212 (𝑋𝑔 𝑐𝑜𝑛), public household demand (𝑋𝑔 𝑔𝑜𝑣 ) and investment demand (𝑋𝑔 𝑖𝑛𝑣). 213 214 Figure 3. Supply and use of commodity g 215 216 217 Where: 218 DPg : domestic production of commodity g. 219 IMg : imports of commodity g. 220 SPg : total supply of the commodity g. 221 DUg : domestic use of the commodity g. 222 EXg : exports of commodity g. 223 INb,g : intermediate input demand of commodity g in industry b. 224 Xg con: private household demand of commodity g. 225 Xg gov: public household demand of commodity g. 226 Domestic production and imports Domestic supply Domestic use and exports 𝐷𝑈𝑔 = ∑b∈B (IN b,g+ Xg con + Xg gov + Xg inv ) DUg SPg (CES/CET) DPg = ∑𝑏=1 𝐵 𝑌𝑌𝑏,𝑔 IMg EXg 12 Xg con: investment demand of commodity g. 227 228 Source: Authors` elaboration 229 230 The model includes one private household that supplies labour and capital to the industries and 231 receives income in return. Capital and labour are assumed to be imperfectly mobile between 232 industries. We also assume one aggregated public household (i.e., government). Consumption 233 of commodity g by the private and public household follows from maximising a CES direct 234 utility function given an income constraint. The CES utility function used implies an income 235 elasticity of one. In addition, as a consumer, the public household imposes taxes and 236 redistributes income. A fixed share of both private and public household income is saved. 237 Savings together with the (minus) surplus on the balance of trade equal investment. Investment 238 demand is modelled using a Leontief production function implying that the demand for an 239 individual commodity is proportional to total investment (Komen & Peerlings, 2001). 240 The model also includes greenhouse gas emissions that are proportionally linked to the 241 production of an industry (𝑌𝑏). 242 243 3.2 Taxes, subsidies, and technological change 244 All transactions in the model can be potentially taxed or subsidised. Taxes can be divided into 245 product and non-product taxes (including subsidies). The latter are levied on income, the first 246 on transactions of commodities. Product taxes drive a price wedge between the demand and the 247 supply price. In the model, we use ad valorem taxes on demand (see Equation 1). 248 249 𝑃𝑑𝑒𝑚𝑎𝑛𝑑 = (1 + 𝑡𝑎𝑥𝑟𝑎𝑡𝑒)𝑃𝑠𝑢𝑝𝑝𝑙𝑦 (1) 250 251 13 A tax increases the price demanders must pay and decreases the price suppliers receive; a 252 subsidy does the opposite. 253 This paper also examines the effects of Hicks neutral technological change in the biobased 254 energy industry. Hicks neutral technological change implies that with the same level and ratio 255 between (all) inputs, more biobased energy can be produced. Equation 2 shows a CES input 256 demand function. In the case of a Hicks neutral technological change, the exogenous scale 257 parameter 𝛤 increases. We include the Hicks neutral technological change in the aggregate 258 demand functions (see Figure 2) of the biobased energy industry. Hicks neutral technological 259 change implies that input demand (𝑥𝑛) and cost of production decrease given a level of output 260 ceteris paribus. However, in the AGE model, the ceteris paribus assumption does not hold, as 261 Hicks neutral technological change lowers the price of wooden biomass as less is needed to 262 produce biobased energy, making it more attractive to demand. Therefore, technological change 263 in the production of biobased energy leads to an increase in the demand for wooden biomass 264 and a lower price for biobased energy. In addition, the lower price of biobased energy leads, 265 because of substitution, to a reduction in the demand for non-biobased energy by all demanders. 266 The degree of substitution between biobased energy and non-biobased energy depends on the 267 degree of substitution (i.e. substitution elasticity) between both in the different industries. 268 𝑥𝑛(. . ) = 𝑦. 𝛤−1. 𝛼𝑛 𝜎. 𝑤𝑛 −𝜎 . (∑ 𝛼𝑛 𝜎𝑁 𝑛=1 . 𝑤𝑛 1−𝜎) 𝜎 1−𝜎 n = 1,..,N (2) 269 where: 𝑥𝑛 - conditional demand for input n, 𝑦 - output, 𝑤𝑛 − price of input n, 𝜎- substitution 270 elasticity, 𝛤 - scale parameter and 𝛼𝑛 - distribution coefficient of input n. 271 272 4. Data 273 The model uses the supply and use tables (SUT) at the basic prices from Latvia`s Central 274 Statistical Bureau for 2015 (CSB Latvia, 2016). A supply table shows in its columns the supply 275 of commodities by the different industries and by imports in an economy for a given period. A 276 14 use table shows in its columns the use of commodities by type of use. Therefore, the use table 277 reveals in its columns the input structure of each industry and the demand for individual 278 commodities by the different final demand categories (Eurostat, 2008). Due to a lack of data, 279 some industries and commodities are aggregated. The data set used in modelling contains 60 280 commodities and 60 industries (see Appendices B and C). 281 Furthermore, we use Latvia’s ‘energy SUT’ in terajoules from Eurostat’s 2015 Physical Energy 282 Flow accounts (Eurostat, 2021) with energy commodities supplied/used by industries to split 283 commodity Electricity, gas, steam and air-conditioning of the SUT in non-biobased and 284 biobased Electricity, gas, steam and air-conditioning, respectively. We consider energy 285 commodities wood, wood waste and other solid biomass, liquid fuels, and biogas as biobased 286 energy commodities. Other energy commodities are fossil-based or renewable energy sources 287 that are not biobased. This implies, for example, that electricity and heat are non-biobased 288 energy products, but they can be produced using both the biobased and non-biobased product. 289 The two energy commodities are used to calculate the shares of biobased and non-biobased 290 energy commodities in the commodity and industry ‘Electricity, gas, steam and air-291 conditioning’ in the SUT, respectively. According to the data of Physical Energy Flow 292 accounts, 7.47% of the total energy supplied and 5.88% of the total energy used comes from 293 biobased energy commodities. 294 295 5. Scenarios and Results 296 5.1 Scenarios 297 In the Base scenario, the model calculates back the actual situation of the Latvian economy in 298 2015. This includes a product tax of 17% for all energy commodities – biobased and non-299 biobased – since all energy producers pay the tax. 300 301 15 Scenario I 302 In Scenario I, we introduce an arbitrary 25% tax on non-biobased energy demand and a subsidy 303 of 10% for biobased energy demand replacing the 17% tax on both products in the Base scenario 304 (see Equation 1 and equations A.29-A.33 in Appendix A). 305 306 Scenario II 307 In Scenario II, a 25% Hicks neutral technological change in the biobased energy production 308 industry is introduced in the Base scenario, where the scale parameter 𝛤 in eq. (2) is increased 309 by 25%. This scenario reflects the technological change in new technologies producing 310 biobased energy that is partially stimulated by government investment from EU funds. The 25% 311 is selected because it leads to a similar increase in the production of biobased energy as in 312 Scenario 1. 313 314 5.2 Results 315 Table 3 shows the outcomes of both scenarios. It is important to note that all price changes are 316 relative to the price numeraire chosen, which is the exchange rate in our case. 317 318 Scenario I 319 Table 3 shows that due to the switch to the subsidy (10%) on biobased energy, its production 320 increases with 57.8%. Table 3 also shows that due to the tax (25%) on non-biobased energy, 321 the price of non-biobased energy increases for buyers (8.3%). Moreover, production in the non-322 biobased energy industry decreases (-6.0%). This leads to a reduction in the value added (-323 2.9%) of this industry. Non-biobased energy is substituted by biobased energy in all industries 324 where the degree of substitution depends on the substitution elasticity between non-biobased 325 and biobased energy. In the model, we assume that this substitution elasticity is large (𝜎 = 1.5), 326 implying that the degree of substitution is large (see Appendix D). In all industries, we see 327 16 therefore a reduction in the demand of non-biobased energy and an increase in the use of 328 biobased energy. Overall, the use of energy falls between 1-3% (a reduction of AENb; see 329 Figure 2). The increase in biobased energy production (57.8%) leads to a reduction in imports 330 of biobased energy products (i.e. natural gas and oil) of 4.2%, making Latvia less dependent on 331 energy imports. However, the subsidy on non-biobased energy increases the imports of this 332 product (64.4%). However, these imports are still very small. Table 3 shows a 1.7% reduction 333 in CO2 emissions assuming CO2 emissions from the biobased energy product to be zero (i.e. 334 being climate neutral). The reduction largely follows from the reduced use of non-biobased 335 energy products (5.7%). Despite this reduction, the target of 50% energy from biobased sources 336 is not reached. 337 Overall, there is a welfare gain (63.5 million euros) in Scenario I, where we measured welfare 338 as private, public, and investment demand changes in prices of the base year. The welfare gain 339 results from a reduction in already existing distortions by introducing the subsidy for biobased 340 products (replacing the 17% tax) and increasing the tax on non-biobased products (from 17% 341 to 25%). Therefore, there is a double dividend. However, whether the double dividend exists 342 depends on the level of tax and subsidy. Sensitivity analyses show that larger taxes and 343 subsidies create welfare losses and that especially the subsidy helps to reduce already existing 344 distortions. Table 3 shows that the increase in tax revenue from the product-related tax on non-345 biobased energy production (117.4 million euro) is larger than the cost of the switch from the 346 tax to the subsidy for the biobased product (45.6 million euro). 347 348 Scenario II 349 Table 3 shows that Hicks neutral technological change of 25% (Scenario II) results in a 350 reduction in the price of biobased energy (-28.1%) and therefore, an increase in production 351 (57.9%) and value added (10.3%) in the biobased energy industry. This leads to a substitution 352 away from non-biobased energy and a reduction in the production (-1.2%) and value added (-353 17 0.8%) in the non-biobased energy industry. Also, in this scenario, import of the biobased energy 354 product fall (-2.5%). The lower price of the biobased energy product decreases imports of the 355 biobased energy product (-12.4%). Again, imports are very small. Compared to Scenario I, the 356 reduction in CO2 emissions is smaller (0.3% versus 1.7%) because the price and production of 357 non-biobased products changes less, and therefore, less substitution takes place. The welfare 358 increase is similar to the welfare increase in Scenario I (61.9 million euros). This welfare gain 359 results from the fact that fewer inputs are needed in the production of biobased energy. This 360 shows the attractiveness of technological change. However, in this scenario, technological 361 change is ‘free’, and this is, of course, not true. 362 Overall, one can conclude that the effects for the Latvian economy are not large. Important 363 reasons for this are the fact that the biobased energy industry is small and even a large growth 364 in production (57.8% and 57.9% in Scenario I and II respectively) does not create a substantial 365 change. Another reason is that in the AGE model factor inputs are mobile between industries 366 making that labour and capital moving out of industries affected negatively by the scenarios 367 can be used elsewhere in the economy leading there to higher production and value added. 368 Finally, the AGE model allows for substitution because of relative price change, again 369 smoothing the effect for the economy as a whole. 370 371 Table 3. Scenario results compared to initial values (i.e., Base scenario) 372 Initial values Scenario I: Tax and subsidy (% change) Scenario II: Technological change (% change) Production* and value added** in million euro of the base year Production of non-biobased energy industry (𝑌𝑏) 1,766.3 -6.0 -1.2 18 Production of biobased energy industry (𝑌𝑏) 110.3 57.8 57.9 Forestry production (𝑌𝑏) 938.9 -0.3 -0.1 Value added in non-biobased energy industry (𝐴𝑃𝑅𝑏) 663.6 -2.9 -0.8 Value added in biobased energy industry (𝐴𝑃𝑅𝑏) 34.2 27.7 10.3 Value added in forestry (𝐴𝑃𝑅𝑏) 352.8 -0.2 -0.1 Prices (index, so no unit) Price of non-biobased energy production (price of 𝐷𝑃𝑔) 1.00 1.2 -0.8 Price of biobased energy production (price of 𝐷𝑃𝑔) 1.00 9.4 -28.1 Price of non-biobased energy demand (price of 𝐷𝑈𝑔) 1.17 8.3 -0.8 Price of biobased energy demand (price of 𝐷𝑈𝑔) 1.17 -16.1 -28.8 CO2 emissions*** in 1000 tons CO2 emissions in non-biobased energy industry 1,757,841 -6.0 -1.2 Total CO2 emissions 6,937,629 -1.7 -0.3 Tax revenue in million euro (nominal) (M euro) (M euro) Product tax paid on non-biobased energy product 293.7 411.1 287.7 Product tax paid on biobased product 23.7 -21.9 24.5 Welfare in million euro of the base year (M euro) (M euro) Laspeyers index 63.5 61.9 * Note: quantities are expressed in million euros for the base year 2015. This implies that initial supply prices 373 (indices) are equal to 1; the initial price for energy demanders is equal to 1.17 due to the 17% tax on energy 374 demand. 375 ** Value added equals the value of capital and labour. 376 19 *** Excluding CO2 emissions from biobased energy commodities that are assumed to be climate neutral. 377 Source: Authors` elaboration 378 379 6. Conclusions and Discussion 380 This paper aims to assess the effect of a tax on non-biobased energy demand and a subsidy on 381 biobased energy demand replacing a lower tax on both to facilitate the substitution of the first 382 for the latter as a step in the transition of Latvia’s economy towards climate neutrality and 383 energy self-sufficiency. Furthermore, the effects of Hicks neutral technological change in the 384 biobased energy industry are examined. The paper uses an applied general equilibrium (AGE) 385 model to assess the effects of a tax, subsidy, and technological change given the expected 386 economy-wide effects and interest in national emissions and welfare. 387 The paper finds that a tax in combination with the subsidy indeed has the expected effects. The 388 Green Deal proposed decarbonisation of the economy by transitioning from fossil-based to 389 renewable sources in energy production. Latvia aims to increase the share of renewable energy 390 production in total energy production to 50% by 2030 (it was 39% in 2017), focussing on the 391 use of wooden biomass in energy production. According to the results of the model, the supply 392 of the biobased energy commodity has increased by 57.8%. However, measures are insufficient 393 to deliver the target, climate neutrality, and energy self-sufficiency. However, there is an overall 394 welfare gain in both Scenario I and Scenario II. So, Scenario I reduces existing distortions in 395 Latvia’s economy (i.e., double dividend). Scenario II (technological change in the biobased 396 energy industry) leads to a similar increase in the production and use of biobased energy. 397 Because in Scenario II the prices of non-biobased energy are affected less than in Scenario I, 398 it`s production and use fall less leading to a lower reduction in CO2 emissions. Although the 399 welfare gain is similar to the gain in Scenario I, Scenario II ignores the costs of technological 400 change and does not explicitly include the incentives needed to implement it. 401 20 To our knowledge, there are no prior studies on the use of wooden biomass for bioenergy 402 production in Latvia. While there are studies of the AGE model on energy taxes, they are not 403 recent. For example, Komen and Peerlings (1999) analysed the effect of an energy tax on small 404 users in the Netherlands using 1990 data. They only find a double dividend in the case of small 405 tax rates. Goulder (1995) discusses the double dividend in more detail, coming to the same 406 conclusion. Welfare effects found by Komen and Peerlings (1999) are also small, like in our 407 case. This is largely due to the substitution possibilities economy-wide and fixed endowments 408 of labour and capital in combination with factor mobility in the models used. 409 This study has three main caveats. First, Latvia devised an action plan in 2019, and is currently 410 undergoing upgrading procedure, although the exact measures are largely unknown. Therefore, 411 it is not possible to calculate the effects of actual policies. This research can contribute to the 412 formulation of such policies. Second, data on energy use and supply are largely aggregated and 413 had to be disaggregated for this study. This involved arbitrary choices. This emphasises the 414 importance of data collection. Related to this, in the base year there is hardly and use of other 415 renewable energy sources than biomass from forests and wood. Wind and solar energy are 416 negligible, although there is some hydropower. It is to be expected that the share of wind and 417 solar energy will grow, requiring that in future research they must be considered separate energy 418 products. Finally, an AGE model is a powerful tool to analyse the economy-wide effects of 419 policies but also comes at a price. For example, the level of aggregation is high, for example, it 420 distinguishes not between, e.g. electricity and heat production. Despite these caveats, this study 421 contributes to the discussion of the transition of the Latvia`s economy towards climate 422 neutrality and energy self-sufficiency. 423 21 References 424 AM - Ministry of Agriculture. (2022). Latvian Forest Sector in Facts and Figures 2022. 425 Available at: https://www.zm.gov.lv/lv/media/1722/download?attachment (Accessed 1 426 April 2025). 427 Arrow, K. J., Chenery, H. B., Minhas, B. S., & Solow, R. M. (1961). Capital-Labor Substitution 428 and Economic Efficiency. The Review of Economics and Statistics, 43(3), 225–250. 429 Bergman, L. (1990). Energy and environmental constraints on growth: A CGE modeling 430 approach. Journal of Policy Modeling, 12(4), 671–691. 431 CSB - Central Statistics Bureau of Latvia. (2021a). ENB010m Electricity production, imports, 432 exports and consumption (mln. kWh). Available at: https://stat.gov.lv/en/statistics-433 themes/business-sectors/energy/tables/enb010m-electricity-production-imports-exports-434 and 435 CSB - Central Statistics Bureau of Latvia. (2021b). ENB020m Imports and consumption of 436 natural gas. Available at: https://stat.gov.lv/en/statistics-themes/business-437 sectors/energy/tables/enb020m-imports-and-consumption-natural-gas 438 CSB - Central Statistics Bureau of Latvia. (2021c). ENB030m Imports, exports and 439 consumption of oil products (thsd tonnes). Available at: https://stat.gov.lv/en/statistics-440 themes/business-sectors/energy/tables/enb030m-imports-exports-and-consumption-oil 441 CSB - Central Statistics Bureau of Latvia. (2021d). MSG040. Forest stand age structure on 1st 442 Jaunary. Available at: 443 https://data.stat.gov.lv/pxweb/en/OSP_OD/OSP_OD__mezsaimn__plat_mez/MSG040.p444 x/ 445 CSB Latvia. (2016). Gross domestic product Supply-Use and Input-Output tables. Available 446 at: https://www.csb.gov.lv/en/statistics/statistics-by-theme/economy/GDP/IOT 447 De Mooij, R. A. (2002). The double dividend of an environmental tax reform. Handbook of 448 Environmental and Resource Economics, 293–306. 449 22 Dervis, K., de Melo, J., & Robinson, S. (1982). General equilibrium models for development 450 policy. Cambridge: Cambridge University Press. 451 EC - European Commission. (n.d.). EU Bioeconomy Monitoring System Dashboard. Available 452 at: https://knowledge4policy.ec.europa.eu/bioeconomy/monitoring_en (Accessed 1 April 453 2025). 454 EC - European Commission b. (n.d.). Bioeconomy&European Green Deal. Available at: 455 https://knowledge4policy.ec.europa.eu/bioeconomy/bioeconomy-european-green-456 deal_en (Accessed 15 January 2022). 457 EM - Ministry of Economics. (2019). National Energy and Climate Plan 2021 – 2030. 458 Available at: https://www.em.gov.lv/lv/nacionalais-energetikas-un-klimata-plans 459 (Accessed 1 April 2025). 460 EU Commission. (2022). REPowerEU: Joint European Action for more affordable, secure and 461 sustainable energy, COM(2022) 108 final. Available at: https://eur-lex.europa.eu/legal-462 content/EN/TXT/?uri=COM%3A2022%3A108%3AFIN (Accessed 1 April 2025). 463 Eurostat. (2008). Eurostat Manual of Supply, Use and Input-Output Tables. Available at: 464 https://ec.europa.eu/eurostat/documents/3859598/5902113/KS-RA-07-013-465 EN.PDF/b0b3d71e-3930-4442-94be-70b36cea9b39 (Accessed 1 April 2025). 466 Eurostat. (2014). Physical Energy Flow Accounts (PEFA) manual. Available at: 467 https://ec.europa.eu/eurostat/documents/1798247/6191537/PEFA-Manual-2014-468 v20140515.pdf 469 Eurostat. (2021). Energy supply and use by NACE Rev.2 activity [env_ac_pefasu]. Available 470 at: https://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=env_ac_pefasu&lang=en 471 FM-Ministry of Finances. (n.d.). Valdība atbalsta 2021. – 2027. gada plānošanas perioda ES 472 fondu vadības likumprojektu. Available at: https://www.cfla.gov.lv/lv/jaunums/valdiba-473 atbalsta-2021-2027-gada-planosanas-perioda-es-fondu-vadibas-likumprojektu (Accessed 474 1 April 2025). 475 23 Fonji, S. F., & Taff, G. N. (2014). Using satellite data to monitor land-use land-cover change 476 in North-eastern Latvia. SpringerPlus, 3(1), 61. 477 Goulder, L. H. (1995), Environmental Taxation and the “Double Dividend”: A Reader's 478 Guide. International Tax and Public Finance 2(2), 157–183. 479 Gunning, J. W., & Keyzer, M. (1995). Applied general equilibrium models for policy analysis 480 (H. Chenery & T. N. Srinivasan, Eds.). Available at: 481 https://econpapers.repec.org/RePEc:eee:devchp:3-35 (Accessed 1 April 2025). 482 Hetemäki, L., Hanewinkel, M., Muys, B., Ollikainen, M., Palahí, M., Trasobares, A., … 483 Potoćnik, J. (2017). Leading the way to a European circular bioeconomy strategy (Vol. 484 5). European Forest Institute. Available at: 485 https://efi.int/sites/default/files/files/publication-bank/2018/efi_fstp_5_2017.pdf 486 (Accessed 1 April 2025). 487 Hurmekoski, E., Jonsson, R., Korhonen, J., Jänis, J., Mäkinen, M., Leskinen, P., & Hetemäki, 488 L. (2018). Diversification of the forest industries: role of new wood-based products. 489 Canadian Journal of Forest Research, 48(12). 490 Komen, M.H.C.; Peerlings, J. H. M. (1999). Energy Taxes in the Netherlands: What are the 491 Dividends? Environmental and Resource Economics, 14(2), 243–268. 492 Komen, M.H.C. and J.H.M. Peerlings (2001). Endogenous technology in dairy farming under 493 environmental restrictions. European Review of Agricultural Economics 28/2: 117-142. 494 Kröger, M., & Raitio, K. (2017). Finnish forest policy in the era of bioeconomy: A pathway to 495 sustainability? Forest Policy and Economics, 77, 6–15. 496 Latvian State Forest Research Institute Silava. (2017). INFORMATION ON LULUCF 497 ACTIONS IN LATVIA Progress report under EU Decision 529/2013/EU Article 10, page 498 10 section 3. https://www.zm.gov.lv/lv/media/705/download?attachment (Accessed 1 499 April 2025). 500 Lazdiņš, A., Lupiķis, A., Butlers, A., Bārdule, A., Kārkliņa, I., Šņepsts, G., & Donis, J. (2019). 501 Latvia’s national forest accounting plan and proposed forest reference level 2021-2025. 502 https://efi.int/sites/default/files/files/publication-bank/2018/efi_fstp_5_2017.pdf https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.zm.gov.lv%2Flv%2Fmedia%2F705%2Fdownload%3Fattachment&data=05%7C02%7Cjack.peerlings%40wur.nl%7C75b003d5477c44048e2508dd705217a1%7C27d137e5761f4dc1af88d26430abb18f%7C0%7C0%7C638790220079036860%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=HVCSamf35EUoP%2F83MQ%2FIs4cSRLWHtWUURTtOOTxY2rc%3D&reserved=0 24 Available at: https://www.fern.org/fileadmin/uploads/fern/Documents/NFAP_Latvia.pdf 503 (Accessed 1 April 2025). 504 Robinson, S. (1989). “Multisectoral models,” Handbook of Development Economics (1st ed., 505 Vol. 2; Hollis Chenery & T.N. Srinivasan, Ed.). Available at: 506 https://ideas.repec.org/h/eee/devchp/2-18.html (Accessed 1 April 2025). 507 Sato, K. (1967). A Two-Level Constant-Elasticity-of-Substitution Production Function. The 508 Review of Economic Studies, 34(2), 201–218. 509 Wolfson, D. J., & Koopmans, C. C. (1996). Regulatory taxation of fossil fuels: Theory and 510 policy. Ecological Economics, 19(1), 55–65. 511 512 513 25 Appendices 514 Appendix A Model description 515 516 Demand and supply equations 517 Aggregate output in industry b (𝑌𝑏) is composed of a hypothetical aggregate energy input 518 (𝐴𝐸𝑁𝑏), aggregate intermediate input (𝐴𝐼𝑁𝑏) and aggregate factor input (𝐴𝑃𝑅𝑏) according a 519 CES production function with constant returns to scale (see glossary at the end of this Appendix 520 for overview of variables, coefficients and sets). Intermediate inputs g in industry b (𝐼𝑁𝑏,𝑔) are 521 transformed into an aggregate energy and aggregate intermediate input according to the CES 522 production functions with constant returns to scale. Labor (𝑃𝑅𝑏,1) and capital (𝑃𝑅𝑏,2) in 523 industry b are transformed into the aggregate factor input, using a CES production function 524 with constant returns to scale. Cost minimization yields CES demand functions for the 525 aggregate energy (A.1), aggregate intermediate (A.2), aggregate factor (A.3), energy (A.4), 526 intermediate (A.5) and factor inputs (A.6): 527 528 𝐴𝐸𝑁𝑏 = 𝑓𝐴𝐸𝑁𝑏 𝐶𝐸𝑆 (𝑌𝑏 , 𝑊𝐴𝐸𝑁𝑏 , 𝑊𝐴𝐼𝑁𝑏 , 𝑊𝐴𝑃𝑅𝑏) ∀𝑏 ∈ 𝐵 (A.1) 529 530 𝐴𝐼𝑁𝑏 = 𝑓𝐴𝐼𝑁𝑏 𝐶𝐸𝑆 (𝑌𝑏 , 𝑊𝐴𝐸𝑁𝑏 , 𝑊𝐴𝐼𝑁𝑏 , 𝑊𝐴𝑃𝑅𝑏) ∀𝑏 ∈ 𝐵 (A.2) 531 532 𝐴𝑃𝑅𝑏 = 𝑓𝐴𝑃𝑅𝑏 𝐶𝐸𝑆 (𝑌𝑏, 𝑊𝐴𝐸𝑁𝑏 , 𝑊𝐴𝐼𝑁𝑏 , 𝑊𝐴𝑃𝑅𝑏) ∀𝑏 ∈ 𝐵 (A.3) 533 534 𝐼𝑁𝑏,𝑔 = 𝑓𝐸𝑁𝑏,𝑔 𝐶𝐸𝑆 (𝐴𝐸𝑁𝑏 , 𝑾𝑰𝑵𝒈) ∀𝑏 ∈ 𝐵, ∀𝑔 ∈ 𝑆𝑒𝑛 (A.4) 535 536 𝐼𝑁𝑏,𝑔 = 𝑓𝐼𝑁𝑏,𝑔 𝐶𝐸𝑆 (𝐴𝐼𝑁𝑏 , 𝑾𝑰𝑵𝒈) ∀𝑏 ∈ 𝐵, ∀𝑔 ∈ 𝑆𝑚𝑎𝑡 (A.5) 537 538 26 𝑃𝑅𝑏,𝑗 = 𝑓𝑃𝑅𝑏,𝑗 𝐶𝐸𝑆 (𝐴𝑃𝑅𝑏 , 𝑾𝑷𝑹𝒃,𝒋) ∀𝑏 ∈ 𝐵, ∀𝑗 ∈ 𝐽 (A.6) 539 540 Supply of output g by industry b (𝑌𝑌𝑏,𝑔) is proportional to the aggregate output (𝑌𝑏) by industry 541 b (A.7). Aggregation of outputs over industries gives domestic production (𝐷𝑃𝑔) of commodity 542 g (A.8): 543 544 𝑌𝑌𝑏,𝑔 = 𝛿𝑏,𝑔 𝑌 × 𝑌𝑏 ∑ 𝛿𝑏,𝑔 𝑌 𝑔∈𝐺 = 1 ∀𝑏 ∈ 𝐵, ∀𝑔 ∈ 𝐺 (A.7) 545 546 𝐷𝑃𝑔 = ∑ 𝑌𝑌𝑏,𝑔 𝐵 𝑏=1 ∀𝑔 ∈ 𝐺 (A.8) 547 548 Domestic production (𝐷𝑃𝑔) and imports (𝐼𝑀𝑔) are aggregated into total supply of commodity 549 g (𝑆𝑃𝑔) using a CES production function with constant returns to scale. This implies that the 550 Armington assumption is adopted (see Dervis et al., 1982, p.221). The total supply is then 551 divided into domestic use (𝐷𝑈𝑔) and exports (𝐸𝑋𝑔) using a CET product transformation 552 function with constant returns to scale. Cost minimization yields CES demand equations for 553 domestic production (A.9) and imports (A.10) and revenue maximization yields CET supply 554 equations for domestic use (A.11) and exports (A.12): 555 556 𝐷𝑃𝑔 = 𝑓𝐷𝑃𝑔 𝐶𝐸𝑆(𝑆𝑃𝑔, 𝑊𝐷𝑃𝑔 , 𝑊𝐼𝑀𝑔) ∀𝑔 ∈ 𝐺 (A.9) 557 558 𝐼𝑀𝑔 = 𝑓𝐼𝑀𝑔 𝐶𝐸𝑆(𝑆𝑃𝑔, 𝑊𝐷𝑃𝑔 , 𝑊𝐼𝑀𝑔) ∀𝑔 ∈ 𝐺 (A.10) 559 560 𝐷𝑈𝑔 = 𝑓𝐷𝑈𝑔 𝐶𝐸𝑇(𝑆𝑃𝑔, 𝑊𝐷𝑈𝑔, 𝑊𝐸𝑋𝑔) ∀𝑔 ∈ 𝐺 (A.11) 561 562 𝐸𝑋𝑔 = 𝑓𝐸𝑋𝑔 𝐶𝐸𝑇(𝑆𝑃𝑔, 𝑊𝐷𝑈𝑔, 𝑊𝐸𝑋𝑔) ∀𝑔 ∈ 𝐺 (A.12) 563 27 564 Total labour (j=1) and total capital (j=2) available in the economy (𝑇𝑃𝑅𝑗) are divided into 565 supply of labour (𝑃𝑅𝑏,1) and capital (𝑃𝑅𝑏,2) by industry using CET product transformation 566 functions with constant returns to scale. Revenue maximisation yields supply functions for 567 labor and capital, respectively (A.13): 568 569 𝑃𝑅𝑏,𝑗 = 𝑓𝑝𝑟𝑏,𝑗 𝐶𝐸𝑇 (𝑇𝑃𝑅𝑗 , 𝑾𝑷𝑹𝒃,𝒋) ∀𝑏 ∈ 𝐵, ∀𝑗 ∈ 𝐽 (A.13) 570 571 Maximization of the CES utility functions yields CES demand equations for the private 572 household (A.14) and public household (A.15): 573 574 𝐶𝑂𝑁𝑔 = 𝑓𝑐𝑜𝑛𝑔 𝐶𝐸𝑆 (𝐸𝑋𝑃𝑐𝑜𝑛, 𝑾𝑪𝑶𝑵𝒈) ∀𝑔 ∈ 𝐺 (A.14) 575 576 𝐺𝑂𝑉𝑔 = 𝑓𝑔𝑜𝑣𝑔 𝐶𝐸𝑆 (𝐸𝑋𝑃𝑔𝑜𝑣, 𝑾𝑮𝑶𝑽𝒈) ∀𝑔 ∈ 𝐺 (A.15) 577 578 The demand for investment goods (𝑋𝑔 𝑖𝑛𝑣) is given by (A.16): 579 580 𝐼𝑁𝑉𝑔 = 𝛿𝑔 𝑖𝑛𝑣 × 𝐼𝑁𝑉 ∑ 𝛿𝑔 𝑖𝑛𝑣 = 1𝑔∈𝑔 ∀𝑔 ∈ 𝐺 (A.16) 581 582 Zero-profit conditions 583 The value of the disaggregated outputs by industry is equal to the value of the aggregate output 584 produced by industry (A.17). The value of aggregate output equals the value of the aggregate 585 energy input, aggregate intermediate input and aggregate factor input (A.18): 586 587 ∑ 𝑊𝐷𝑃𝑔𝑔∈𝐺 𝑌𝑏,𝑔 = 𝑊𝑌𝑏 × 𝑌𝑏 ∀𝑏 ∈ 𝐵 (A.17) 588 28 589 𝑊𝑌𝑏𝑌𝑏 = 𝑊𝐴𝐸𝑁𝑏 × 𝐴𝐸𝑁𝑏 + 𝑊𝐴𝐼𝑁𝑏 × 𝐴𝐼𝑁𝑏 + 𝑊𝐴𝑃𝑅𝑏 × 𝐴𝑃𝑅𝑏 ∀𝑏 ∈ 𝐵 (A.18) 590 591 The value of the aggregate energy input is equal to the value of the energy inputs by industry 592 (A.19). The value of the aggregate intermediate input equals the value of the intermediate inputs 593 by industry (A.20). The value of the aggregate factor input equals the value of labour and capital 594 by industry (A.21): 595 596 𝑊𝐴𝐸𝑁𝑏 × 𝐴𝐸𝑁𝑏 = ∑ 𝑊𝐼𝑁𝑏,𝑔 × 𝐼𝑁𝑏,𝑔𝑔∈𝑆𝑒𝑛 ∀𝑏 ∈ 𝐵 (A.19) 597 598 𝑊𝐴𝐼𝑁𝑏 × 𝐴𝐼𝑁𝑏 = ∑ 𝑊𝐼𝑁𝑏,𝑔 × 𝐼𝑁𝑏,𝑔𝑔∈𝑆𝑚𝑎𝑡 ∀𝑏 ∈ 𝐵 (A.20) 599 600 𝑊𝑁𝐴𝑃𝑅𝑏 × 𝐴𝑃𝑅𝑏 = ∑ 𝑊𝑃𝑅𝑏,𝑗 × 𝑃𝑅𝑏,𝑗 2 𝑗=1 ∀𝑏 ∈ 𝐵 (A.21) 601 602 The value of total supply (𝑆𝑃𝑔) equals the sum of the value of domestic production and imports 603 (A.22) and the sum of the value of domestic use and exports by commodity (A.23): 604 605 𝑊𝑆𝑃𝑔 × 𝑆𝑃𝑔 = 𝑊𝐷𝑃𝑔 × 𝐷𝑃𝑔 + 𝑊𝐼𝑀𝑔 × 𝐼𝑀𝑔 ∀𝑔 ∈ 𝐺 (A.22) 606 607 𝑊𝑆𝑃𝑔 × 𝑆𝑃𝑔 = 𝑊𝐷𝑈𝑔 × 𝐷𝑈𝑔 + 𝑊𝐸𝑋𝑔 × 𝐸𝑋𝑔 ∀𝑔 ∈ 𝐺 (A.23) 608 609 The value of the supply of labour and capital equals the value of the total availability of labour 610 and capital, respectively (A.24). 611 612 ∑ (𝑊𝑃𝑅𝑏,𝑗 × 𝑃𝑅𝑏,𝑗)𝐵∈𝑏 = 𝑊𝑇𝑃𝑅𝑗 × 𝑇𝑃𝑅𝑗 ∀𝑗 ∈ 𝐽 (A.24) 613 614 29 The value of the demand for individual investment goods equals the expenditure on investment 615 (A.25): 616 617 ∑ (𝑊𝐼𝑁𝑉𝑔 × 𝐼𝑁𝑉𝑔)𝑔∈𝐺 = 𝑊𝐼𝑁𝑉 × 𝐼𝑁𝑉 (A.25) 618 619 Margins 620 The value of demand/supply for margins per commodity (𝑀𝐴𝑅𝑔) is assumed to form a share of 621 the value of the different demand categories and is given by (A.26). 622 623 𝑀𝐴𝑅𝑔 = 𝑚𝑔 𝑠𝑝 × 𝑊𝐷𝑈𝑔 × 𝐷𝑈𝑔 + 𝑚𝑔 𝑠𝑝 × 𝑊𝐸𝑋𝑔 × 𝐸𝑋𝑔 Gg (A.26) 624 625 The total value of the margins is zero (A.27): 626 627 ∑ 𝑀𝐴𝑅𝑔𝑔∈𝐺 = 0 (A.27) 628 629 Equilibrium conditions for commodities 630 Total domestic use equals intermediate, private household, public household and investment 631 demand (A.28): 632 633 𝐷𝑈𝑔 = ∑ 𝐼𝑁𝑏,𝑔 + 𝐶𝑂𝑁𝑔 + 𝐺𝑂𝑉𝑔 + 𝐼𝑁𝑉𝑔𝑏∈𝐵 Gg (A.28) 634 635 Price equations 636 Indirect taxes and margins drive a wedge between the demanders’ and suppliers’ price of 637 domestic use. This applies to commodities consumed by the private household (A.29), the 638 public household (A.30), as investment commodities (A.31), as intermediate inputs (A.32) and 639 as exports (A.33): 640 30 641 𝑊𝐶𝑂𝑁𝑔 = (1 + 𝑚𝑔 𝑠𝑝 + 𝑡𝑔 𝑠𝑝) × 𝑊𝐷𝑈𝑔 (A.29) 642 643 𝑊𝐺𝑂𝑉𝑔 = (1 + 𝑚𝑔 𝑠𝑝 + 𝑡𝑔 𝑠𝑝) × 𝑊𝐷𝑈𝑔 (A.30) 644 645 𝑊𝐼𝑁𝑉𝑔 = (1 + 𝑚𝑔 𝑠𝑝 + 𝑡𝑔 𝑠𝑝) × 𝑊𝐷𝑈𝑔 (A.31) 646 647 𝑊𝐼𝑁𝑔 = (1 + 𝑚𝑔 𝑠𝑝 + 𝑡𝑔 𝑠𝑝) × 𝑊𝐷𝑈𝑔 (A.32) 648 649 𝑊𝐺𝐸𝑋𝑔 = (1 + 𝑚𝑔 𝑠𝑝 + 𝑡𝑔 𝑠𝑝) × 𝑊𝐸𝑋𝑔 (A.33) 650 651 The price received for exports (A.34) and paid for imports (A.35) are equal to the world market 652 price times the exchange rate: 653 654 𝑊𝐺𝐸𝑋𝑔 = 𝑊𝑃𝐸𝑋𝑔 ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ × 𝐸𝑅̅̅ ̅̅ (A.34) 655 𝑊𝐼𝑀𝑔 = 𝑊𝑃𝐼𝑀𝑔 ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ × 𝐸𝑅̅̅ ̅̅ (A.35) 656 657 Non-product related taxes are levied on the total value of labour and capital used by industry 658 (A.36): 659 660 𝑊𝐴𝑃𝑅𝑏 = (1 + 𝑡𝑏 𝑎𝑝𝑟) × 𝑊𝑁𝐴𝑃𝑅𝑏 (A.36) 661 662 Income formation and distribution 663 Tax revenue from product (𝑃𝑇𝑋) and non-product related taxes (𝑁𝑃𝑇𝑋) and social contribution 664 (𝑆𝑂𝐶𝐶𝑂𝑁) is given by equations A.37, A.38 and A.39, respectively: 665 31 666 𝑃𝑇𝑋 = ∑ (𝑡𝑔 𝑠𝑝 × 𝑊𝐷𝑈𝑔 × 𝐷𝑈𝑔 + 𝑡𝑔 𝑠𝑝 × 𝑊𝐸𝑋𝑔 × 𝐸𝑋𝑔)𝑔∈𝑆 (A.37) 667 668 𝑁𝑃𝑇𝑋 = ∑ (𝑡𝑏 𝑎𝑝𝑟 × 𝑊𝑁𝐴𝑃𝑅𝑏 × 𝐴𝑃𝑅𝑏)𝑏∈𝐵 (A.38) 669 670 𝑆𝑂𝐶𝐶𝑂𝑁 = 𝑟𝑠𝑜𝑐𝑐𝑜𝑛 × 𝐿𝐴𝐵𝐼 (A.39) 671 672 Labour income (𝐿𝐴𝐵𝐼), capital income (𝐶𝐴𝑃𝐼) and capital depreciation (DEP) are given by 673 equation A.40, A.41 and A.42 respectively: 674 675 𝐿𝐴𝐵𝐼 = 𝑊𝑇𝑃𝑅1 × 𝑇𝑃𝑅1 (A.40) 676 677 𝐶𝐴𝑃𝐼 = 𝑊𝑇𝑃𝑅2 × 𝑇𝑃𝑅2 (A.41) 678 679 𝐷𝐸𝑃 = 𝑟𝑐𝑎𝑝𝑑𝑒𝑝 × 𝐶𝐴𝑃𝐼 (A.42) 680 681 The income (𝐼𝑐𝑜𝑛), expenditure 𝐸𝑋𝑃𝑐𝑜𝑛 and savings (𝑆𝐴𝑉𝑐𝑜𝑛) of the private household are 682 given by equations A.43, A.44 and A.45 respectively: 683 684 𝐼𝑐𝑜𝑛 = (1 − 𝑟𝑠𝑜𝑐𝑐𝑜𝑛) × 𝐿𝐴𝐵𝐼 + (1 − 𝑟𝑐𝑎𝑝𝑑𝑒𝑝) × 𝐶𝐴𝑃𝐼 (A.43) 685 686 𝐸𝑋𝑃𝑐𝑜𝑛 = (1 − 𝑠𝑐𝑜𝑛) × 𝐼𝑐𝑜𝑛 (A.44) 687 688 𝑆𝐴𝑉𝑐𝑜𝑛 = 𝑠𝑐𝑜𝑛 × 𝐼𝑐𝑜𝑛 (A.45) 689 690 32 The income (𝐼𝑔𝑜𝑣), expenditure (𝐸𝑋𝑃𝑔𝑜𝑣) and savings (𝑆𝐴𝑉𝑔𝑜𝑣) of the public household are 691 given by equations A.46, A.47 and A.48, respectively: 692 693 𝐼𝑔𝑜𝑣 = 𝑃𝑇𝑋 + 𝑁𝑃𝑇𝑋 + 𝑆𝑂𝐶𝐶𝑂𝑁 (A.46) 694 695 𝐸𝑋𝑃𝑔𝑜𝑣 = (1 − 𝑠𝑔𝑜𝑣) × 𝐼𝑔𝑜𝑣 (A.47) 696 697 𝑆𝐴𝑉𝑔𝑜𝑣 = 𝑠𝑔𝑜𝑣 × 𝐼𝑔𝑜𝑣 (A.48) 698 699 Total savings (A.49) equal the sum of the savings of the private household, public household 700 and capital depreciation minus the surplus on the balance of trade (BBAR). As the balance of 701 trade is expressed in foreign price it must be multiplied by the exchange rate (ER): 702 703 𝑆𝐴𝑉 = 𝑆𝐴𝑉𝑐𝑜𝑛 + 𝑆𝐴𝑉𝑔𝑜𝑣 + 𝐷𝐸𝑃 − 𝐵𝐵𝐴𝑅 × 𝐸𝑅̅̅ ̅̅ (A.49) 704 705 The surplus on the balance of trade (A.50) equals: 706 707 𝐵𝐵𝐴𝑅 = ∑ (𝑊𝑃𝐸𝑋̅̅ ̅̅ ̅̅ ̅̅ �̅� × 𝐸𝑋𝑔)𝑔∈𝐺 − ∑ (𝑊𝑃𝐼𝑀̅̅ ̅̅ ̅̅ ̅̅ �̅� × 𝐼𝑀𝑔)𝑔∈𝐺 (A.50) 708 709 The value of investment (𝐼𝑁𝑉𝑇) equals the value of the savings (A.51) but also the value of the 710 investment commodities (A.52). Due to Walras’ law, one equation must be dropped to keep the 711 model identified. In the operational model this is Equation A.52. 712 713 𝐼𝑁𝑉𝑇 = 𝑆𝐴𝑉 (A.51) 714 715 𝐼𝑁𝑉𝑇 = 𝑊𝐼𝑁𝑉 × 𝐼𝑁𝑉 (A.52) 716 33 717 Price numéraire 718 An applied general equilibrium model is homogenous of degree zero. This model selects the 719 exchange rate as price numéraire. 720 721 Environment 722 CO2, CH4, N2O emissions of an industry (𝐸𝑀𝐼𝑆𝑏) equals the share (amount) of emissions 723 (𝛿𝑏 𝑒𝑚𝑖𝑠) emitted by the industry(𝑌𝑏 𝑜𝑙𝑑). 724 725 𝛿𝑏 𝑒𝑚𝑖𝑠 = 𝐸𝑀𝐼𝑆𝑏 𝑜𝑙𝑑/𝑌𝑏 𝑜𝑙𝑑 (A.53) 726 727 𝐸𝑀𝐼𝑆𝑏 = 𝛿𝑏 𝑒𝑚𝑖𝑠 × 𝑌𝑏 (A.54) 728 729 Welfare change 730 As a welfare measure, the Laspeyres index of real income change is taken. This index compares 731 commodity bundles between two equilibria (e.g. before and after a policy change), using the 732 prices of the initial equilibrium (A.55). This welfare measure allows for the calculation of the 733 welfare effects of savings other than the private savings of which the underlying optimizing 734 behaviour is not modelled explicitly (i.e., capital depreciation, government deficit and the 735 balance of trade). Since savings are equal to investments, the bundle of investment commodities 736 represent welfare derived from saving. 737 738 𝑊𝐸𝐿𝐹 = 739 ∑ (𝑊𝐶𝑂𝑁𝑔 𝑜𝑙𝑑 × 𝐶𝑂𝑁𝑔 𝑜𝑙𝑑)𝑔∈𝑆 +∑ (𝑊𝐺𝑂𝑉𝑔 𝑜𝑙𝑑 × 𝐺𝑂𝑉𝑔 𝑜𝑙𝑑)𝑔∈𝑆 + ∑ (𝑊𝐼𝑁𝑉𝑔 𝑜𝑙𝑑 × 𝐼𝑁𝑉𝑔 𝑜𝑙𝑑) −𝑔∈𝑆 740 ∑ (𝑊𝐶𝑂𝑁𝑔 𝑜𝑙𝑑 × 𝐶𝑂𝑁𝑔 ) −𝑔∈𝑆 ∑ (𝑊𝐺𝑂𝑉𝑔 𝑜𝑙𝑑 × 𝐺𝑂𝑉𝑔 )𝑔∈𝑆 − ∑ (𝑊𝐼𝑁𝑉𝑔 𝑜𝑙𝑑 × 𝐼𝑁𝑉𝑔 )𝑔∈𝑆 (A.55) 741 742 34 Glossary 743 Variables: 744 Quantities 745 𝑌𝑏 aggregate output of industry b 746 𝐴𝐸𝑁𝑏 aggregate intermediate energy input in industry b 747 𝐴𝐼𝑁𝑏 aggregate intermediate materials input in industry b 748 𝐴𝑃𝑅𝑏 aggregate factor input in industry b 749 𝐼𝑁𝑏,𝑔 intermediate input g in industry b 750 𝑃𝑅𝑏,𝑗 labour (j=1) and capital (j=2) in industry b 751 𝑌𝑌𝑏,𝑔 output g of industry b 752 𝐷𝑃𝑔 domestic production of commodity g 753 𝐼𝑀𝑔 import of commodity g 754 𝑆𝑃𝑔 total supply of commodity g 755 𝐷𝑈𝑔 domestic use of commodity g 756 𝐸𝑋𝑔 export of commodity g 757 𝑇𝑃𝑅𝑗 ̅̅ ̅̅ ̅̅ ̅ labour (j=1) and capital (j=2) endowment in the economy 758 𝐶𝑂𝑁𝑔 consumer demand for commodity g 759 𝐺𝑂𝑉𝑔 government demand for commodity g 760 𝐼𝑁𝑉𝑔 demand of investment good g 761 𝐼𝑁𝑉 aggregate investment good 762 𝐸𝑀𝐼𝑆𝑏 emissions of industry b 763 764 Prices (quantity symbols plus a W) 765 𝑊𝐴𝐸𝑁𝑏 price of aggregate intermediate energy input in industry b 766 𝑊𝐴𝐼𝑁𝑏 price of aggregate intermediate materials input in industry b 767 35 𝑊𝐴𝑃𝑅𝑏 price of aggregate factor input including tax in industry b 768 𝑊𝑁𝐴𝑃𝑅𝑏 price of aggregate factor input excluding the tax in industry b 769 𝑊𝐼𝑁𝑔 price of intermediate input g 770 WPRb,j price of labour (j=1) or capital (j=2) used in industry b 771 𝑊𝐼𝑀𝑔 price of import of commodity g 772 𝑊𝑃𝐼𝑀𝑔 ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ world market price of import of commodity g 773 𝑊𝐷𝑃𝑔 price of domestic production of commodity g 774 𝑊𝑌𝑏 price of aggregate output of industry b 775 𝑊𝑆𝑃𝑔 price of total supply of commodity g 776 𝑊𝐷𝑈𝑔 price of domestic use of commodity g 777 𝑊𝐸𝑋𝑔 price of export excluding taxes and margins of commodity g 778 𝑊𝐺𝐸𝑋𝑔 price of export including taxes and margins of commodity g 779 𝑊𝑃𝐸𝑋𝑔 ̅̅ ̅̅ ̅̅ ̅̅ ̅̅ world market price of export of commodity g 780 𝑊𝐶𝑂𝑁𝑔 price of commodity g demanded by the private household 781 𝑊𝐺𝑂𝑉𝑔 price of commodity g demanded by the public household 782 𝑊𝑇𝑃𝑅𝑗 price of labour (j=1) and capital (j=2) endowment in the economy 783 𝑊𝐼𝑁𝑉𝑔 price of investment good g 784 𝑊𝐼𝑁𝑉 price of aggregate investment good 785 786 Other 787 𝐸𝑅̅̅ ̅̅ exchange rate 788 𝑀𝐴𝑅𝑔 value of demand/supply of margins by commodity g 789 𝐿𝐴𝐵𝐼 labour income 790 𝐶𝐴𝑃𝐼 capital income 791 𝐷𝐸𝑃 capital depreciation 792 36 𝐼𝑐𝑜𝑛 income of private household 793 𝐼𝑔𝑜𝑣 income of public household 794 𝐸𝑋𝑃𝑐𝑜𝑛 expenditure of private household 795 𝐸𝑋𝑃𝑔𝑜𝑣 expenditure of public household 796 𝑆𝐴𝑉𝑐𝑜𝑛 expenditure of private household 797 𝑆𝐴𝑉𝑔𝑜𝑣 expenditure of public household 798 𝑆𝐴𝑉 total savings 799 𝐵𝐵𝐴𝑅 surplus on the trade balance (in foreign prices) 800 𝐷𝐸𝑃 capital depreciation 801 𝐼𝑁𝑉𝑇 value of the investment 802 𝑃𝑇𝑋 value of the product related taxes 803 𝑁𝑃𝑇𝑋 value of the non-product related taxes 804 𝑆𝑂𝐶𝐶𝑂𝑁 social contributions 805 WELF welfare change 806 807 Fixed coefficients 808 𝛿𝑏 𝑒𝑚𝑖𝑠 coefficient dividing aggregate emissions by industry b 809 𝑚𝑔 𝑠𝑝 margins for commodity g 810 𝑡𝑔 𝑠𝑝 tax rate for product related taxes on commodity g 811 𝑡𝑏 𝑎𝑝𝑟 tax rate for non-product related taxes in industry b 812 𝛿𝑏,𝑔 𝑌 coefficient dividing aggregate output into outputs g in industry b 813 𝛿𝑔 𝑖𝑛𝑣 coefficient dividing aggregate investment good into investment good g 814 𝑠𝑐𝑜𝑛 savings rate of the private household 815 𝑠𝑔𝑜𝑣 savings rate of the public household 816 𝑟𝑠𝑜𝑐𝑐𝑜𝑛 share of labour income that goes to social contributions 817 37 𝑟𝑐𝑎𝑝𝑑𝑒𝑝 capital depreciation as share of capital income 818 819 Sets and subsets: 820 G: goods, g = 1 to 60 821 B: industries, b = 1 to 60 822 J: factors, j = 1 (labour) j = 2 (capital) 823 :GSen  subset energy commodities: g = 24,25 824 :GSmat  subset materials: g = 1,…,23,26,…,60 825 826 Miscellaneous 827 gov = public household; con = private household; inv = investment; en = energy; old = base 828 year value; CES = Constant Elasticity Substitution; CET = Constant Elasticity of 829 Transformation 830 831 Bold printed variables represent a vector; variables with a bar represent exogenous variables. 832 833 Appendix B Aggregation level of commodities and industries in Supply-Use tables 834 CPA/NACE code Commodity/Industry A01 Crop and animal production, hunting and related service activities A02 Forestry and logging A03 Fishing and aquaculture B Mining and quarrying C10_12 Manufacture of food products, beverages and tobacco products C13_15 Manufacture of textiles, wearing apparel and leather products 38 C16 Manufacture of wood and of products of wood and cork, except furniture; manufacture of articles of straw and plaiting materials C17 Manufacture of paper and paper products C18 Printing and reproduction of recorded media C19 Manufacture of coke and refined petroleum products C20 Manufacture of chemicals and chemical products C21 Manufacture of basic pharmaceutical products and pharmaceutical preparations C22 Manufacture of rubber and plastic products C23 Manufacture of other non-metallic mineral products C24 Manufacture of basic metals C25 Manufacture of fabricated metal products, except machinery and equipment C26 Manufacture of computer, electronic and optical products C27 Manufacture of electrical equipment C28 Manufacture of machinery and equipment n.e.c. C29 Manufacture of motor vehicles, trailers and semi-trailers C30 Manufacture of other transport equipment C31_32 Manufacture of furniture; other manufacturing C33 Repair and installation of machinery and equipment D35-1 Biobased energy D35-2 Non-biobased energy E36 Water collection, treatment and supply E37_39 Sewerage; waste collection, treatment and disposal activities; materials recovery; remediation activities and other waste management services F Construction 39 G Wholesale and retail trade H49 Land transport and transport via pipelines H50_51 Water and Air transport H52 Warehousing and support activities for transportation H53 Postal and courier activities I Accommodation and food service activities J58 Publishing activities J59_60 Motion picture, video and television program production, sound recording and music publishing activities; programming and broadcasting activities J61 Telecommunications J62_63 Computer programming, consultancy and related activities; information service activities K64 Financial service activities, except insurance and pension funding K65 Insurance, reinsurance and pension funding, except compulsory social security K66 Activities auxiliary to financial services and insurance activities L68 Real estate activities M69_70 Legal and accounting activities; activities of head offices; management consultancy activities M71 Architectural and engineering activities; technical testing and analysis M72 Scientific research and development M73 Advertising and market research M74_75 Other professional, scientific and technical activities; veterinary activities N77 Rental and leasing activities N78 Employment activities 40 N79 Travel agency, tour operator reservation service and related activities N80_82 Security and investigation activities; services to buildings and landscape activities; office administrative, office support and other business support activities O Public administration and defense; compulsory social security P Education Q86 Human health activities Q87_88 Social work activities R90_92 Creative, arts and entertainment activities; libraries, archives, museums and other cultural activities; gambling and betting activities R93 Sports activities and amusement and recreation activities S94 Activities of membership organizations S95 Repair of computers and personal and household goods S96_T Other personal service activities 835 Appendix C Industries ranked according to the largest shares in supply and use of biobased 836 energy commodities 837 Suppliers of Energy commodities* Users of Energy commodities** Forestry and logging Manufacture of wood and of products of wood and cork, except furniture; manufacture of articles of straw and plaiting materials Fishing and aquaculture Manufacture of chemicals and chemical products Manufacture of machinery and equipment n.e.c. Residential care activities and social work activities without accommodation 41 Rental and leasing activities Sewerage, waste management, remediation activities Manufacture of wood and of products of wood and cork, except furniture; manufacture of articles of straw and plaiting materials Manufacture of chemicals and chemical products Manufacture of furniture; other manufacturing Retail trade, except of motor vehicles and motorcycles * 90%< 838 **30%< 839 Source: author`s calculations based on Eurostat, 2021 840 841 Appendix D Substitution (sigma) and transformation (omega) elasticities 842 Symbol Function Value SIGMAY (B) CES for output 0.4 SIGMAPR (B) CES for aggregate primary input 0.4 SIGMAE (B) CES for aggregate energy input 1.5 SIGMAX (B) CES for aggregate intermediate input 0.4 SIGMADOMS (G) CES for domestic supply 1.5 OMEGADOMD (G) CET for domestic demand -1.5 843 844