Linear General Equilibrium Model of Energy Demand and CO 2 Emissions Generated By the Andalusian Productive System

Transcription

1 America Joural of Ecoomics ad Busiess Admiistratio 4 (4): , 2012 ISSN: Sciece Publicatio doi: /ajebasp Published Olie 4 (4) 2012 ( Liear Geeral Equilibrium Model of Eergy Demad ad CO 2 Emissios Geerated By the Adalusia Productive System 1,2 Mauel Alejadro Cardeete, 2 Patricia Fuetes Saguar ad 3 Clemete Polo 1 Europea Commissio, Joit Research Cetre-Istitute for Prospective Techological Studies, C/Ica Garcilaso, 3, E Seville, Spai 2 Departmet of Ecoomics, Uiversidad Pablo Olavide, Ctra. Utrera, km.1, s/, E Seville, Spai 3 Departmet of Ecoomics, Uiversidad Autooma de Barceloa, Campus de Bellaterra, Edifici B, E , Barceloa, Spai ABSTRACT I this study we apply a multiplier decompositio methodology of a liear geeral equilibrium model based o the regioal social accoutig matrix to the Adalusia ecoomy. The aim of this methodology is to separate the size of the differet effects i terms of eergy expediture ad total emissios geerated by the whole productive system to satisfy the fial demad of each brach of the Adalusia ecoomy ad the direct emissios geerated to produce eergy for each subsystem. Keywords: CO 2 Emissios, Social Accoutig, Liear Geeral, Multiplier Decompositio Methodology 1. INTRODUCTION The use of eergy commodities by productive activities is the key cause of greehouse gases. May studies coducted over recet decades have addressed this questio i respose to the growig awareess of the eed to stop climate chage. I Adalusia, productive activities geerate 90% of the regio s total emissios, where emissios geerated per uit of eergy cosumed are greater tha the emissios geerated for fial cosumptio. This is attributable to the large price discrimiatio withi this sector. As the emissios geerated by the productive system have the biggest share i total emissios, this is priority target for the adoptio of measures with a view to achievig the emissio targets established i the Kyoto Protocol. For this reaso, this study focuses o aalysig the emissios geerated by the Adalusia productive subsystems i order to establish the resposibilities of each oe for satisfyig the fial demads for productio for a more detailed aalysis of the subsystems i the iput-output framework, see (Sraffa, 1960; Heimler, 1991; Sachez-Choliz ad Duarte, 2003). The accoutig multipliers calculated for each subsystem are disaggregated ito four partial effects that we ca use to decompose total emissios. The statistical iformatio used to specify the SAM model parameters is the 2000 Adalusia social accoutig matrix (SAMAND00) at basic prices ad vector C of CO 2 emissios. The study is structured as follows. I the ext sectio we describe the methodology used to decompose sectoral productio ito differet effects. Later we exted this methodology by mappig productio to CO 2 emissios. I the ext sectio, we apply the methodology to the Adalusia productive system. Fially, we state the coclusios ad discuss the costraits ad possible system. The productive system is divided ito several extesios of the model. Correspodig Author: Mauel Alejadro Cardeete, Departmet of Ecoomics, Uiversidad Pablo Olavide, Ctra. Utrera, km.1, s/, E Seville, Spai Tel: (+34)

2 2. LINEAR SAM MODEL-BASED MULTIPLIER DECOMPOSITION I this sectio, we preset a additive accoutig multiplier decompositio methodology Multiplicative accoutig multipliers (Polo et al., 1990) are aother meas of decompositio that is equally capable of disaggregatig a ecoomy s reveue geeratio process based o the SAMAND00. This methodology ca disaggregate the output of the subsystems (i.e., groupigs of productive sectors) ito differet effects. Based o a classificatio of istitutios as edogeous ad m exogeous, the reveue of the edogeous istitutios satisfies Equatio 1: y = Ay + Amy m = Ay + d (1) where, y ad y m deote the vector of the reveue of the edogeous ad exogeous istitutios, respectively; A ad A m are two ad m matrices defied by the coefficiets of the expediture of the edogeous ad exogeous istitutios, respectively, destied for the edogeous istitutios; ad d = Amy m is the vector of the exogeous reveue destied for the edogeous istitutios. As i the Leotief model, the reveue of the edogeous istitutios ca be obtaied as Equatio 2: y ( ) -1 = I - A d = My d (2) where, M is the geeralized multiplier matrix. Substitutig (2) ito (1), we get: y = AMy + d (3) Followig Alcatara ad Padilla (2008) Whereas Alcatara ad Padilla (2007) aalyse the services subsystem, we exted the aalysis to the other subsystems makig up the Adalusia productive system, if we deote the subsystem of edogeous accouts uder aalysis by s ad the complemetary subset by r ad partitio the expediture ad multiplier matrices Equatio 4a ad 4b: A A A 0 0 A A = = + A A 0 A A 0 ss sr ss sr rs rr rr rs (4a) y A 0 0 A = + y 0 A A 0 Mss Msr ds ds + M M d d s ss sr R rr rs rs rr r r (5) where, y s ad y r are the colum vectors of the accouts that subsystem S icludes or excludes, respectively ad d s ad d r are the vectors of exogeous reveue destied for subsystem s or the complemetary subsystem r, respectively. Supposig that exogeous reveue destied for the complemetary subsystem r is zero, d r = 0, we get the followig system of Equatio 6: y = A M d + A M d + d s ss ss s sr rs s s y = A M d + A M d r rr rs s rs ss s (6) This system provides the reveue of the istitutios whe the demad destied for the complemetary subsystem is zero. The reveue of the istitutios icluded i subsystem s is decomposed ito three summads: A ss M ss d s represets the reveue geerated withi the subsystem s itself whe it receives exogeous reveue d s. It is called ow effect ad deoted y OE s A sr M rs d s represets reveue geerated i subsystem s due to the icrease i reveue geerated i the complemetary subsystem r to satisfy the exogeous demad d s. It is called the feedback effect ad is deoted y FBE s d s represets the direct reveue of subsystem s whe the exogeous reveue is d s. It is called the scale effect (SCE) ad is deoted y SCE s The secod Equatio (5) provides the reveue geerated i the complemetary subsystem r whe the exogeous reveue is d s. The secod term, A sr M ss d s, represets the reveue iduced i the complemetary subsystem r by the reveue geerated i the istitutios s; ad the first term, A rr M rs d s, reflects the reveue iduced i subsystem r by the geeratio of reveue i its istitutios iduced by the exogeous reveue destied for the istitutios i s. It is called the spill-over effect ad is deoted y SOE r. M M ss sr M = M rs M rr Equatio (3) ca be writte as Equatio 5: (4b) 3. AN ESTIMATE OF SECTOR CO 2 EMISSIONS To lik the above aalysis with CO 2 emissios, we eed a vector that coverts the moetary uits of the 217

3 model ito emissio uits. To do this, we use the vector c e of emissios per uit of each eergy commodity used (Table 11). Multiplyig this vector by the submatrix A E, we get a vector of uit emissios for each istitutio Equatio 7: c = c' A (7) E where, c is the traspose of the vector. By dividig this vector ito two, oe for the istitutios i subsystem r, c r ad aother for subsystem s, we get a breakdow of total CO 2 emissios Equatio 8: T E = c' sy s + c' ry r (8) Geerated to satisfy the vector of fial demad d s by the four above-metioed effects: Emissios associated with the ow effect Equatio 9: OE OE E = c' sy s = c' sassmssd s (9) Emissios associated with the feedback effect Equatio 10: FBE FBE E = c' sy s = c' sasrmrssd s (10) Emissios associated with the scale effect Equatio 11: SCE y s = c' sd s (11) Emissios associated with the spill-over effect Equatio 12: ( ) E = c' y = c' A M d + A M d (12) SOE s r r r rr rs s rs ss s To get the emissios by braches, we have to diagoalize vector d ( d ), whereby, for scale effect emissios, we get the vector em S SCE (1 s) Equatio 13: em = c,d SCE S S S (13) where, these are the emissios geerated by each brach of subsystem s to produce the uits destied to satisfy its fial demad. We get the spill-over effect i terms of emissios by calculatig the vector em S SOE (1 R) Equatio 14: em = c A M + A M d ( ) SOE S r rr rs rs ss S (14) 218 This vector is composed of emissios geerated by the other productive braches to produce what each brach of subsystem s demads to be able to satisfy its fial demad. To get the emissios geerated by the productio of the aalysed subsystem destied to meet the eeds of each subsystem brach for ow iputs to satisfy its fial demad (emissios due to the ow effect), we calculate the vector em S OE (1 s) Equatio 15: em = c A M d OE S S SS SS S (15) I this case, the results suggest that the fial demad of each brach of the subsystem has a pull effect o the emissios that it geerates. Similarly, we calculate the emissios due to the feedback effect Equatio 16: em = c A M d, (16) FBE S S sr rs S where, each elemet of this vector represets the emissios geerated by the subsystem as a whole to produce what the other braches belogig to r eed to be able to produce what each brach of subsystem s demads to satisfy its fial demad. Therefore, the total effect (EM TE ) i terms of emissios directly or idirectly geerated to satisfy the fial demad of subsystem s would be Equatio 17: EM TE = EM SCE + EM OEP + EM FBE + EMSOE (17) ad the total effect i terms of emissios due to the fial demad of each brach of s would be the vector em S TE (1 s) Equatio 18: em = em + em + em + em. (18) TE SCE OE FBE SOE S S S S S Moreover, we ca use Equatio (5) agai, this time makig the vector D S zero, to get the sales made by subsystem s to the other sectors to eable them to satisfy their ow fial demads. Followig the same procedure as above, we would get for the other braches Note that this equatio reflects the spill-over effect for braches of r i the same way as (9) does for the braches of s Equatio 19: A M d + A M d + 0 = Y (19) r ss sr r sr rr r S If we multiply this expressio by the vector of emissios, we will have the emissios geerated by sales

4 to other sectors to satisfy their demad, which we will deote Sales Effect (SE) Equatio 20: EM = c Y (20) r SE S S ad the emissios for each brach of the subsystem by sales to other sectors would be Equatio 21: em = c Yˆ (21) SE r S S S I a SAM aalysis such as this, the emissios due to the sales effect accout for the ifluece of the edogeized accouts, as we will see later. This meas that these emissios are closer tha i the iput-output model to the direct emissios, called EM DE, geerated by each productive subsystem. We calculate EM DE by multiplyig the uit emissios geerated per uit produced by each subsystem by subsystem productio Equatio 22: EM DE = c S Y S (22) 4. RESULTS AND DISCUSSION I accordace with the Ster Review classificatio Ster (2007), we divide the Adalusia productive system ito six groups, called subsystems, show i Table 1, alogside their SAMAND00 (Table 12) for Adalusia SAMAND00 accouts ad iput-output (MIOAND00) framework equivaleces for the year 2000 equivaleces. I the followig, we apply the multiplier decompositio methodology outlied to the SAMAND00 usig a SAM model (Table 10) shows the results of applyig the multiplier decompositio methodology to the SAMAND00 iput-output model, that is, without edogeizig labour, capital ad cosumptio (cosiderig the labour (29), capital (30) ad cosumptio (31) accouts as edogeous) for each subsystem. This outputs the productio of each subsystem destied to satisfy the fial demad of each subsystem brach divided ito the differet effects formig what we term the total effect. We trasform these results ito CO 2 emissios as explaied before. This way, as show i Table 2 below, we get the emissios geerated to satisfy the fial demad of each subsystem. Table 1. Productive subsystems of the adalusia ecoomy SAMAND00 accout equivalece Subsystem 1: Primary 1, 2 ad 3 Subsystem 2: Eergy 4, 5, 7, 8 ad 9 Subsystem 3: Idustry 6, 10 to 21 Subsystem 4: Costructio 22 Subsystem 5: Trasport 25 ad commuicatios Subsystem 6: Services 23, 24, 26, 27 ad 28 Source: Ow elaboratio Fig. 1. Direct ad total emissios of the Adalusia ecoomy s productive subsystems over sectoral emissios (%) 219

5 Table 2. Total ad direct CO 2 emissios geerated by the Adalusia productive system (2000) based o the SAM* model Primary Eergy Idustry Costructio Trasport ad Services (1) (2) (3) (4) commuicatios (5) (6) EM OE EM FBE EM SCE , , , EM SOE 1, , , , EM TE 2, , , , , , %EM TE /EM AE ** 6.70% 5.36% 21.50% 24.69% 2.92% 27.95% EM SE 1, , , , , EM DE 2, , , , , , %EM DE /EM AE ** 5.64% 47.57% 16.68% 8.22% 13.80% 8.08% (*) Stated i kilotoes (kt) of CO 2. (**) Adalusia Ecoomy Sectoral Emissios (EMAE): 41,616 Source: Ow elaboratio. Table 3. Total ad direct uit emissios of the adalusia productive system Primary (1) Eergy (2) Idustry (3) Costructio (4) Trasport ad commuicatios (5) Services (6) EM TE /D EM DE /Y d * * Yd: Domestic productio; Source: Ow elaboratio Table 4. CO 2 emissios by braches of the primary subsystem Arable farmig , , , ,944.3 Livestock farmig Fishery TOTAL SUBSYSTEM , , , ,348.7 Source: Ow elaboratio Table 5. CO 2 emissios by braches of the eergy subsystem Coal Oil ad atural gas Oil refiery , , , ,004.8 Electricity , ,258.4 Gas TOTAL SUBSYSTEM , , , ,797.3 Source: Ow elaboratio Table 2 shows the differet effects i EM TE for each subsystem as a whole. The results are disaggregated by braches (Table 4-9) ad cotaiig the emissios due to the ow effect (EM OE ), that is, emissios geerated by subsystem s to produce what the subsystem itself requires to satisfy its fial demad. This is followed by the feedback effect (EM FBE ), that is, emissios geerated by subsystem s to produce what it eeds to sell to other sectors for them to produce what subsystem s demads from them to satisfy its fial demad. The ext row shows the scale effect (EM SCE ), which are the emissios geerated by the subsystem s to produce what it is to sell directly to its fial demad ad 220 fially, we have the spill-over effect (EM SOE ), that is, the emissios geerated by the other sectors to produce what subsystem s demads from them to be able to satisfy its fial demad. The come the emissios due to sales made by each subsystem to the other sectors r (EM SE ) i order to satisfy the fial demads of r ad fially, the emissios geerated directly by the productio of each subsystem, which we call the direct effect (EM DE ). Fially, Table 2 lists both the EM TE ad the EM DE for each subsystem, as a ratio of total sectoral emissios (which we call EM AE ) of the Adalusia ecoomy.

6 Table 6. CO 2 emissios by braches of the idustry subsystem Other extractive Water Food , , Textile ad leather Wood workig Chemical idustry , , , ,061.8 Miig ad iro ad steel , ,312.2 Metal workig Machiery Vehicle Buildig materials ,022.7 Trasport Other maufacturig TOTAL SUBSYSTEM , , , , ,939.8 Source: Ow elaboratio Table 7. CO 2 emissios by braches of the costructio subsystem Costructio , , , ,422.8 TOTAL SUBSYSTEM , , , ,422.8 Source: Ow elaboratio These ratios are also illustrated as percetages i Fig. 1 to give a clearer ad fuller picture of the results ad make the detailed aalysis that follows easier to uderstad ad follow. As Fig. 1 ad Table 2 show, the results split the subsystems ito two groups. O the oe had, we have subsystems 1, 3, 4 ad 6, which have higher EM TE tha EM DE, whereas the opposite applies to subsystems 2 ad 5. Startig with the first of these groups, the primary (1), idustry (3), costructio (4) ad services (6) subsystems are sectors that have a pull effect o emissios geerated by the ecoomy, that is, they geerate high idirect emissios to satisfy their fial demad. Cosequetly, their EM TE are greater tha the emissios that are geerated directly by sector productio (EM DE ). Geerally speakig, a characteristic of these subsystems is that they have high EM SOE ad low EM SE. The services (6) subsystem, whose EM SOE accout for 24.12% of total Adalusia productive system emissios, raks top of this group, due to the emissios geerated by productio i other sectors ad required by the services subsystem to satisfy its ow fial demad. It is this that explais the services subsystem s high level of EM TE, accoutig for over a quarter of the emissios of all the productive activities. These results reflect the sizeable pull effect of this sector, ot usually cosidered as a major polluter, o emissios geerated by the Adalusia ecoomy Alcatara ad Padilla (2008). 221 O the other had, EM SE are the secod raked emissios i this sector, albeit well below EM SOE. This is because they cotai sales for private cosumptio. For this reaso, EM SCE are lower tha i the iput-output model (Table 10), where they rak secod for this subsystem ad although agai cosiderably lower tha EM SOE, are relatively high compared with the other subsystems. Fially, this sector has strikigly high EM FBE levels compared with the other subsystems, oce agai showig how importat the services sector is withi the Adalusia ecoomy because of its high iterdepedecies with the other subsystems. Lookig at the results by braches we have coducted a study of whole subsystems rather tha a brach-by-brach aalysis. For this reaso, the results stated by braches refer to emissios geerated to satisfy the fial demad of the subsystem to which they belog. For a brach-by-brach aalysis, the ow effect has to be split ito two (a itra-brach effect ad a iter-brach effect), Alcatara ad Padilla (2008). We fid that the services ot for sale sector (28) raks top withi this subsystem. This sector has the highest level of EM TE at about 20% of the total emissios of the productive activities, primarily due to EM SOE. Aother oteworthy poit is that the trade (24) brach, except fuel trade (23), behaves differetly withi this subsystem, as it accouts for over 50% of the subsystem s EM DE, whereas EM TE are relatively small.

7 Table 8. CO 2 emissios by braches of the trasport ad commuicatios subsystem Trasport ad commuicatios , , ,742.6 TOTAL SUBSYSTEM , , ,742.6 Source: Ow elaboratio Table 9. CO 2 emissios by braches of the services subsystem Fuel trade Other trade , ,773.6 Other services , , Services for sale , , Services ot for sale , , TOTAL SUBSYSTEM , , , ,364.4 Source: Ow elaboratio Table 10. Total ad direct CO 2 emissios geerated by the Adalusia productive system (2000) based o the iput-output model* Subsystem 1 Subsystem 2 Subsystem 3 Subsystem 4 Subsystem 5 Subsystem 6 EM OE , , EM FBE EM SCE 1, , , , , , EM SOE 1, ,74 4, , , EM TE 2, , , , , , %EM TE /EM AE ** EM SE 1, , , , EM DE 2, , , , , , %EM DE /EM AE ** (*) Stated i kilotoes (kt) of CO 2. (**) Adalusia ecoomy sectoral emissios (EM AE ): 42,616. Source: Ow elaboratio Table 11. Vector C of emissios per uit of eergy commodity used (kt CO 2 / 1000 ) Coal Oil Refiery Electricity Gas Itermediate Demad Fial Demad Source: Ow elaboratio based o Maresa ad Sacho (2004) Comparig the results for this subsystem, we fid that although there are some differeces betwee the two models, they both show that the services subsystem has a sizeable pull effect o the emissios of the other braches. Note firstly that there is big a differece betwee the two models as regards both EM SOE ad cosequetly, EM TE. Aother strikig poit is that the emissios due to the spill-over effect i the services for sale sector (27) ad especially, the services ot for sale sector (28) are greater i the SAM model. These differeces agai highlight the fact that SAM models accout for relatioships ot captured by iputoutput models. 222 Rakig secod i terms of emissios i this first group is the costructio subsystem (4). This subsystem behaves similarly to the services subsystem ad its EM SOE are agai high, due to this sector s massive demads for eergy ad idustrial iputs. I this case, there is o eed to aalyse this sector by braches, as it is composed of a sigle SAMAND00 accout (22). Although this subsystem s direct emissios are uder 9%, this is a relatively high value cosiderig that it is a sigle SAMAND00 accout. Aother oteworthy poit is the high level of EM TE of costructio i Adalusia, which accouts for early 25% of the emissios of the whole productive system, eve though it is a sigle accout ad raks above other subsystems like idustry (3).

8 Table 12. Sectoral structure of SAMAND00 ad MIOAND00 equivaleces SAMAND00 MIOAND00 1. Arable farmig 1 to 3 2. Livestock framig 4 ad 5 3. Fishery, fish farmig ad related activities 6 4. Eergy extractive 7 5. Other extractive 8 ad 9 6. Oil refiery ad uclear waste treatmet Electrical eergy productio ad distributio Gas, water vapour ad hot water productio Water collectio, treatmet ad distributio Food 10 to Textile ad leather 20 to Wood workig 23 ad Chemicals 27 ad Miig ad iro ad steel Metal workig Machiery 35 to Vehicles Buildig materials 30 to Trasport 41 ad Other maufacturig 25, 29, 43 to Costructio 49 ad Vehicle ad fuel trade Other trade 52 to Trasport ad commuicatios 57 to Other services 61 to 63, 66 to 71, 73, 83 ad Services for sale 64, 65, 72, 76, 78, 80, 81, 85 ad Services ot for sale 74, 75, 77, 79 ad 82 Source: Ow elaboratio based o MIOAND00 (Adalusia Statistical Office) Comparig these results with the iput-output model outcomes, we fid that this subsystem raks fourth i terms of total emissios, because it has lower EM SOE, although, as applies to other subsystems, the fidigs are similar ad more proouced i the SAM. The ext highest raked subsystem i this group is idustry (3). I this case, the differeces betwee EM TE (21.50%) ad EM DE (16.68%) are less, albeit sigificat i both cases. This was to be expected takig ito accout that this is the largest subsystem with a total of 13 braches. Agai, the explaatio is to be foud i the EM SOE, which accout for over 60% of EM TE. Compared with the other subsystems, idustry has the greatest EM SCE ad EM OE of the Adalusia productive system ad the secod highest value for EM FBE. The explaatio is uquestioably that idustrial iputs are very importat for all productive systems, idustry icluded, as well as for fial demad, especially ivestmet. Fially, idustry s high EM SE accout for the fact that the differeces betwee its EM TE ad its EM DE are less tha for the services subsystem. 223 As regards the results by braches, the highest rakig sectors are the food (11), chemicals (14), miig ad iro ad steel (15) ad buildig materials (19) idustries, although they behave differetly i terms of EM DE ad EM TE. The chemicals sector (14) has the greatest EM DE values, accoutig for 7.36% of total sectoral emissios ad rakig secod i terms of EM TE, after food (11). The other two braches whose productio geerates most direct emissios withi this subsystem are miig ad iro ad steel (15) ad buildig materials (19). Lookig at EM TE, though, we fid that the food idustry (11) geerates the highest values, with relatively low EM DE, but EM SOE that accout for just over 80% of its EM TE ad which are much higher tha its EM DE. It is followed by the chemicals (14) ad miig ad iro ad steel (15) idustries, whereas the buildig materials sector (19) has low EM TE levels. These results highlight the role played by the food idustry i the Adalusia ecoomy, where it is a key sector ad therefore has a big ifluece o the pollutat emissios geerated by the system.

9 The lowest rakig subsystem i this first group is the primary sector (1), which is the subsystem that geerates the fewest emissios as a whole i terms of EM DE ad raks fourth, with slightly higher EM TE, above the eergy (2) ad trasport (5) subsystems. However, a aalysis by shows up sigificat differeces, as the agriculture sector (1) accouts for most of the subsystem emissios, with 80% of both direct ad total emissios. Fially, let us stress that the emissios geerated by this subsystem also have a oteworthy EM SE ad EM SOE compoet. This explais why there is so little differece betwee the total ad direct emissios. Focusig ow o the secod group, we have subsystems 2 ad 5. I both, EM DE are greater tha EM TE. This idicates that these are braches that absorb emissios from other sectors ad are characterized by high emissios per sales to other sectors (EM SE ) ad low values for emissios due to the spill-over effect (EM SOE ). Startig with the top-rakig subsystem withi this secod group, we have the eergy subsystem (2). I this case, the high emissios due to the ow effect, scale effect ad sales effect were to be expected. They are explaied by the high use of eergy commodities by both the productive system ad fial demad. As Table 2 shows, the emissios geerated directly by the eergy subsystem (2) i its productio process (EM DE ) accout for almost 50% of total emissios by productive activities, where this subsystem raks well above the rest of the ecoomy. This is explaied primarily by the EM SE. The value of EM SE i this model amouts to almost 90% of this subsystem s direct emissios, as this sector absorbs all the emissios geerated by the eergy eeds of all the other subsystems to satisfy their fial demads. Additioally, we fid that the emissios due to the fial demad of eergy (EM TE ) accout for 5.36% of sectoral emissios, of which just over 70% correspod to EM SCE, that is, to output destied for the fial demad of eergy. I relatio to the braches of this subsystem, we fid that the electricity sector raks top, with direct emissios accoutig for over 39% of the emissios geerated by the productive activities. This is the brach of the Adalusia ecoomy that releases most emissios ito the atmosphere Oe of the reasos for electricity sector beig the biggest polluter of the Adalusia ecoomy is the use of a high percetage of coal productio, this beig the eergy commodity that geerates most emissios whe cosumed. This is explaied by the high emissios geerated by sales to other sectors (EM SE, 224 which accout for 37% of total emissios i the Adalusia productive system). However, the EM TE of this brach accout for just 1.2% of sectoral emissios. This is explaied partly by the model, as we have edogeized private cosumptio, meaig that fial demad (exogeous accouts) does ot iclude cosumptio ad therefore, the emissios due to the scale effect are less tha we would get with a iput-output model, as are emissios due to ow effect, as the demad they have to satisfy is lower. The values of the other effects, especially the spill-over ad feedback effects, are very low for the electricity brach. I the case of coal, whose direct emissios amout to 1.10% of sectoral emissios ad whose EM TE are just uder 0.3%, emissios geerated by sales to other sectors are oteworthy, give that most of its productio is destied for use as iput for the eergy braches, especially, the electricity sector, whereas its fial uses have declied. The EM TE of the refiery brach (7), which accouts for direct emissios amoutig to just over 7% of sectoral emissios, raks top i terms of emissios geerated by this subsystem. This shows just how importat the fial demad of this sector is i terms of emissios. This is likely to be due to exports (where 35% of the output of this sector is destied for this use), as sales to other sectors icludes private trasport I Adalusia s iput-output framework for the year 2000 ad therefore i the SAMAND00, private trasport is allocated to the oil refiery row ad the private cosumptio colum. This explais its high value, which eve outraks total emissios. Oe of the most oteworthy results of the aalysis of this subsystem is the low level of emissios of the gas brach (9), accoutig for less tha 0.2% of total sectoral emissios i both cases (EM TE ad EM DE ), where the values for emissios due to the ow effect ad to the sales effect are the most sigificat. Fially, aother iterestig poit is that EM SOE accout, i all cases, for a very small percetage of sectoral emissios, as expected. Cotiuig with the aalysis of the secod group, the secod raked subsystem after eergy is trasport ad commuicatios (5). The aalysis of this subsystem is similar, as it is a sector of which both the productive activities ad fial demad make a lot of use. Its direct emissios are therefore greater tha emissios due to the total effect ad it has the secod highest value for EM SE, after the eergy subsystem (2). This reflects how this sector absorbs emissios from the rest of the system.

10 Agai, as i the case of costructio, o aalysis by braches is required, as this subsystem is composed of oly oe SAMAND00 accout (25). To complete this aalysis, let us detail the results of this exercise preseted i Table 3. Table 3 shows the emissios due to both the total effect ad the direct effect, weighted, i the first case, agaist the fial demad of each subsystem brach ad i the secod, agaist the domestic output of each brach of the subsystem. The aim here is to fie tue the results, as, i some cases, they are due to a greater weight i the ecoomy ad i others, to sizeable price differeces. The values for EM TE /D idicate the emissios geerated by the system as a whole to be able to satisfy a uit of the system s fial demad, whereas EM DE /Y d are emissios geerated by the subsystem per uit output. Here we fid that the eergy subsystem (2) raks well above all the others i terms of direct emissios, whereas the highest value for total emissios geerated by the system to satisfy oe uit of fial demad is for the trasport ad commuicatios (5) subsystem. Also, comparig the two idicators for all the subsystems, we fid that oly i the case of subsystem 2 is the secod idicator greater tha the first. This idicates that this is the subsystem that absorbs most of the emissios geerated i the system. Noteworthy are the low emissios per uit of demad satisfied i the idustry subsystem. Additioally, whereas the services sector is raked as the subsystem that has the greatest EM TE i the aalysis set out i Table 2, here we fid that the emissios per uit of demad satisfied are the secod lowest across the system, after the idustry subsystem. Eve so, we also fid that there is a big differece betwee these emissios ad emissios geerated per uit of productio i the services subsystem. 5. CONCLUSION I this study we have developed a methodology that is useful for extedig the iformatio about CO 2 emissios by the productive sectors of the Adalusia ecoomy, as, apart from idetifyig the emissios that each brach geerates i its productive process, we are able to ascertai what idirect emissios (geerated by other braches) are ecessary to satisfy the fial demad of each brach. Thaks to the applied methodology, therefore, we ca output the direct emissios geerated by each subsystem ad the total emissios, i.e., direct ad idirect emissios, that are geerated to satisfy the fial demad 225 of each subsystem. We ca also separate these emissios ito differet effects. Calculatig these emissios ca be helpful for detectig which braches ad subsystems are the oes that release most emissios ito the atmosphere ad especially, which are the demads that have the biggest pull effect o emissios geerated i the ecoomy, plus which are the braches ad subsystems most affected by these demads. We have divided the subsystems ito two groups depedig o the results. The first, composed of the primary, eergy, idustry ad costructio subsystems, are characterized by high EM SOE ad low EM SE. O the other had, subsystems 2 ad 5 are characterized by high EM SE ad lower EM SOE. The first group cotais the sectors with a sizeable pull effect o emissios geerated by the system, especially the services subsystem. The secod group icludes sectors that have a high absorptio effect of emissios geerated by the system. I coclusio, the subsystems that have the highest levels of direct emissios are: The eergy subsystem (2), accoutig for 47.6% of total sectoral emissios. The highest raked sector withi this subsystem is electricity (8), accoutig for 39% of EM AE The idustry subsystem (3), which geerates 16.7% of sectoral emissios. The top raked sector i this subsystem is chemicals (14) The trasport ad commuicatios (5) subsystem, which accouts for 13.8% of sectoral emissios These subsystems ad braches therefore have a sizeable absorptio effect of system emissios. The highest rakig subsystems i terms of emissios due to the total effect are: The services subsystem (6), accoutig for 27.95% of total sectoral emissios. Withi this subsystem, the top-rakig braches are trade (24) i terms of direct emissios ad above all, services ot for sale (28) i terms of total emissios The costructio subsystem (4), whose total emissios accout for 25% of EM TE The idustry subsystem (3), which geerates 21.5% of total sectoral emissios. The top-rakig brach withi this subsystem is the food idustry (10) I this case, these are sectors that, as already metioed, have a sizeable pull effect o system emissios.

11 As regards the differeces i the results of applyig this methodology to the SAM model, where we edogeize the labour (29), capital (30) ad cosumptio (31) accouts ad the iput-output model, if we compare Table 2 with Table 10, we see that the fidigs are similar, although more proouced i the case of total emissios i the SAM model. The key differeces betwee the two models are that the scale effect does ot iclude private cosumptio. This is reflected i the sales to other sectors effect ad the spill-over effect, which icludes emissios due to the edogeized accouts. The results of this exercise are potetially useful for extedig the, sometimes deficiet, iformatio about emissios, apart from providig guidace o policies for applicatio i the future. Note, however, that the differece i emissios ca, i some cases, be explaied by the subsystems havig a greater weight i the ecoomy, as show i Table 3, or by sizeable price differeces, as i the case of the eergy subsystem, which is kow to apply sizeable price discrimiatio. I this respect, a iterestig extesio of this research could be to take ito accout such price differeces i both eergy uits ad the fial prices of the differet braches. Additioally, these aalyses should to start to iclude reewable eergies, which are gaiig i importace i ecoomic ad evirometal terms. For this to be possible, the iput-output frameworks would have to separate these activities. Alcatara, V. ad E. Padilla, Iput-output subsystems ad pollutio: A applicatio to the service sector ad CO 2 emissios i Spai. Ecol. Eco., 68: DOI: /j.ecoleco Heimler, A., Likages ad vertical itegratio i the Chiese ecoomy. Rev. Eco. Stat., 73: Maresa, A. ad F. Sacho, Eergy itesities ad CO 2 emissios i cataloia: A SAM aalysis. It. J. Eviro. Workplace Employmet, 1: Polo, C., D. Rolad-host ad F. Sacho, Distribucio de la reta e u Modelo SAM de la ecoomía española. Estadística Española, 32: Sachez-Choliz, J. ad R. Duarte, Aalysig pollutio by way of vertically itegrated coefficiets, with a applicatio to the water sector i Arago. Camb. J. Eco., 27: DOI: /cje/ Sraffa, P., Productio of Commodities by Meas of Commodities. 1st Ed., CUP Archive, Cambridge, ISBN-10: , pp: 98. Ster, N.N.H., The Ecoomics of Climate Chage: The Ster Review. 1st Ed., Cambridge Uiversity Press, Cambridge, ISBN-10: , pp: REFERENCES Alcatara, V. ad E. Padilla, Subsistemas Iput- Output y cotamiacio: Ua aplicacio al Sector Servicios y las Emisioes de CO 2 e España. II Joradas Españolas de Aalisis Iput-Output Zaragoza: Crecimieto, Demada y Recursos aturales. Libro de Comuicacioes. Asociacio Hispaoamericaa de Iput-Output. Zaragoza. 226