The relevance of supply shocks for inflation: the spanish case

Fo Pee Re iew The Rele ance o Supply Shocks o In la ion: The Spanish Case Jou nal: Applied Economics Manusc ip ID: APE-06-0390.R1 Jou nal Selec ion: Applied Economics JEL Code: E31 - P ice Le el|In la ion|De la ion < E3 - P ices, Business Fluc ua ions, and Cycles < E - Mac oeconomics and Mone a y Economics Keywo ds: In la ion, Nominal igidi ies, Supply shocks, Skewness, Spanish Regions Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip Fo Pee Re iew 1 The Rele ance o Supply Shocks o In la ion: The Spanish Case 1. In oduc ion The idea o he p esen con ibu ion is based on se e al ac o s: i) Spain is a coun y cha ac e ised by a pe sis en mode a e in la ion di e en ial wi h he co e EU coun ies –see Eu opean Cen al Bank (2003). ii) The igu es o Spanish in la ion ha e sligh ly inc eased in ecen yea s and he Spanish Go e nmen aces p oblems o con ol in la ion –see Bank o Spain (2006). iii) The i egula e olu ion o oil p ices in ecen yea s, wi h se e al ad e se supply shocks, dese es a lo o in e na ional a en ion –see Kilian (2005). Ou pape ies o shed some ligh join ly on hese ac o s, om he Spanish pe spec i e, p oposing some explana ions mainly based on he use o Ball and Mankiw’s (1994, 1995) app oach. In o de o implemen panel da a echniques and p o ide addi ional in o ma ion a a egional le el, we pay special a en ion o he Spanish egional in la ion da a, al hough we also include in ou analysis se e al con ol a iables. Empi ical e idence shows ha in la ion and he highe momen s o he dis ibu ion o ela i e p ices a e posi i e co ela ed, agains he heo e ical p edic ions o he lexible p ice model. Ball and Mankiw (1994,1995) show ha in la ion is mainly in luenced by skewness, a guing ha , in p esence o nominal igidi ies, due o he ac ha i ms ace menu cos s, changes in he p ice le el and skewness a e posi i ely co ela ed; e ec ha can be magni ied by he s anda d de ia ion o he dis ibu ion, deno ed as ela i e p ice a iabili y (RPV) in his s and o he li e a u e. Ou s udy ies o check i he skewness-in la ion ela ion holds o Spain and i he beha iou o Spanish egions is homogeneous wi h espec o i . The analysis o such ela ion can be ele an in he sense ha hese au ho s show ha skewness is a p oxy o supply shocks, and he e o e ha ela ion is explaining how sensi i e he in la ion is when a supply shock a ec s he economy and i a supply shock a ec s o he same ex en all egions. Posi i e in la ion-skewness and in la ion-RPV ela ions a e suppo ed by he da a, bu esul s a e no conclusi e abou which ela ion is s onge . On one hand, o Page 1 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 2 pe iods wi h an annual in la ion a e lowe han 4%-5%, he in la ion-skewness ela ion is s onge han he in la ion-RPV one –see Ball and Mankiw (1995) o he US, Lou enco and G uen (1995) o Aus alia, Amano and Macklem (1997) o Canada, Auc emanne e al. (2002) o Belgium and Ca aballo and Usabiaga (2004a,b) o Spain, among o he s. Mo eo e , o some high in la ion coun ies he e is e idence o a posi i e associa ion be ween in la ion and skewness, as Ra ai (2004) shows o Hunga y in a pe iod wi h an annual in la ion a e anging om 15% o 30%. On he o he hand, o s udies co e ing pe iods wi h changing in la ion a e, he e idence is mixed. Fo example, Hall and Ya es (1998), o he 1975-1996 pe iod in he Uni ed Kingdom, ind a weake in la ion-skewness ela ion han he in la ion-RPV one o he whole pe iod. Mo e p ecisely, bo h ela ions a e s onge o he high in la ion pe iod and he o me is e en nega i e o he low in la ion pe iod, in con as o he esul s ob ained by Assa sson (2004) and Ca aballo and Dabús (2005). The i s au ho inds o Sweden ha RPV and skewness a e mo e impo an in explaining in la ion in he low in la ion pe iod han in he high in la ion one. Ca aballo and Dabús (2005) ind he same esul s o Assa sson (2004) o skewness bu no o RPV. These au ho s ocus on Spain and A gen ina, concluding ha RPV is signi ican o bo h low and high in la ion pe iods o A gen ina and only o he high in la ion pe iod in Spain. In addi ion, hey ind ha skewness is signi ican o he low in la ion pe iod bu no in he high in la ion pe iod in bo h coun ies, e en hough he mean in la ion a e in each pe iod di e s s ongly ac oss hem. In ac , he mean annual in la ion a e o A gen ina in he low in la ion pe iod (a ound 23%), is highe han he Spanish in la ion a e in he high in la ion pe iod (14%). Finally, Döpke and Pie dzioch (2003), o he 1969-2000 pe iod in Ge many, ind ha bo h ela ions a e posi i e, bu none o hem is clea ly s onge . Table 1 ies o summa ise he main empi ical e idence on his opic. [Table 1] This mixed e idence can be due o di e en easons, and specially o he ac ha he ela ion be ween in la ion and he highe momen s o he dis ibu ion o changes in ela i e p ices is e y sensi i e o changes in he ea u es o mean in la ion. Page 2 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 3 Gene ally in low in la ion coun ies bo h a iables a e signi ican bu depending on he end o in la ion a ela ion can be mo e signi ican han he o he .1 The main con ibu ions o his pape , in compa ison wi h p e ious ones in his a ea o he Spanish economy –Ca aballo and Usabiaga (1994a, 1994b), Ca aballo and Dabús (2005)–, a e he ollowing: we wo k wi h a highe deg ee o disagg ega ion in he da a, a e y impo an ea u e in his kind o li e a u e based on p ice changes dis ibu ion unc ions; we ex end and upda e he pe iod o analysis; and we inco po a e as con ol a iables he main economic a iables ela ed o his opic a ailable o he Spanish economy wi h a mon hly equency. The es o he pape is o ganised as ollows. Sec ion 2 p esen s he main da a and a iables. In sec ion 3 we de elop a p elimina y analysis o he 17 Spanish egions. Sec ion 4 pe o ms a panel da a analysis. In sec ion 5 se e al con ol a iables a e included, and sec ion 6 concludes. 2. Main da a and a iables Ou analysis e e s o he 1993.02-2005.12 pe iod. We a e awa e o he sho ness o his sample pe iod (13 yea s) in compa ison wi h o he s udies, bu i is no possible o ex end i , due o he impo an da a equi emen s o ou analy ical me hodology, wi h a high deg ee o disagg ega ion in he da a, as well as he use o se e al con ol a iables. Only he pe iod conside ed ul ils all his da a ma ching. Howe e , we ha e o ake in o accoun ha he da a a e mon hly, a equency which is no commonly used in he li e a u e, and consequen ly we ge 155 obse a ions o each se ies. Ou sample pe iod can be clea ly di ided in o wo subpe iods. The i s one goes om 1993.02 o 1998.12, and is cha ac e ised by a nega i e end in la ion, and a mean mon hly in la ion a e a ound 0.28%. The second one is he 1999.01-2005.12 pe iod, in which no end in la ion is ound and p esen s a mean mon hly in la ion a e a ound 0.26%. 1 See Ca aballo and Dabús (2005) o u he de ails. Page 3 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 4 The main da a used a e he se ies o mon hly change a es o consume p ice indexes, disagg ega ed by goods and se ices (57 ca ego ies), o he 17 Spanish egions elabo a ed by he Ins i u o Nacional de Es adís ica (INE). The weigh o each subg oup o e ed by he INE is de ined as he p opo ion o expense made on ha a icle in ela ion o o al expendi u e made by households. The weigh is kep cons an by he INE along he 1993.02-2001.12 pe iod, bu since 2002 he e has been a change in he me hodology and he weigh s change e e y yea . This ac is aken in o accoun when he momen s o he dis ibu ion o in la ion a e calcula ed. Ano he change in he me hodology is he in oduc ion o sales in he index. In o de o a oid he p oblems caused by his change, we emo e he seasonal componen using he TRAMO-SEATS me hod. As con ol a iables we use he a e o unemploymen , he indus ial p oduc ion index, he gene al e ail ade index, he shopping mall e ail ade index, he oil p ices and he indus ial p ice index. We p o ide in o ma ion abou hem in he co esponding sec ion. As a as he cons uc ion o he main a iables is conce ned, we use he second and hi d c oss-sec ional momen o he dis ibu ion o p ice changes. The exp essions o he s anda d de ia ion o each egion (RPVj ) and he skewness o each egion (Sj ) a e as ollows: [ ( ) ] 5,0 2 1 ∑ = −= n i j ij ijj wRPV ππ ; [ ] ( ) 3 3 1 j j ij n i ij j S w S ππ − = ∑ = whe e π e e s o in la ion a e, i o goods, j o egions and o ime pe iods. The e o e, π : Spanish in la ion in pe iod ; π j : in la ion o egion j in pe iod ; π ij : in la ion o subg oup i in egion j in pe iod ; and wij is he weigh o each subg oup i and egion j used by INE. 3. In la ion, RPV and skewness: p elimina y analysis on a egional basis In his sec ion a p elimina y egion-by- egion analysis is pe omed. In o de o implemen i , we un he ollowing eg ession o each egion Page 4 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 5 j j j j j j j jj RPVS εββπβαπ ++++= −321,1 [1] The lagged in la ion e m is included in o de o cap u e he pe sis ence o he se ies. Be o e unning he eg essions we ha e checked he s a iona i y o he se ies.2 Fo he 17 egions in la ion p esen s a nega i e de e minis ic end o he 1993-1998 pe iod, bu he e is no end in he 1999-2005 one. This ea u e o in la ion is included in he eg essions. The eg essions a e es ima ed by o dina y leas squa es (OLS).3 As usual, he p- alue o he -s a is ic (in b acke s in he ables) is co ec ed o he e oscedas ici y by means o he Whi e me hod. We show he esul s o each subpe iod (Tables 2 and 3) and o he whole pe iod (Table 4). [Table 2] [Table 3] [Table 4] As i can be seen om he ables, skewness is signi ican in 13 egions o he 1993-1998 subpe iod, in 15 egions o he 1999-2005 subpe iod, and in 13 egions o he whole pe iod, and i s coe icien emains unchanged o he di e en sample pe iods. Howe e , he beha iou o RPV is no so homogeneous ac oss pe iods, and ables show ha i is signi ican in 7 egions o he i s subpe iod, i is no signi ican in any egion o he second subpe iod, and i is signi ican in 8 egions o he whole pe iod (in 6 o hem i was signi ican in he i s pe iod as well), and i s coe icien a ies conside ably among sample pe iods. I is also in e es ing o poin ou he ema kable changes in he adjus ed R2 depending on he pe iod conside ed; he 2 In he Appendix we p esen he esul s o a common uni oo –B ei ung (2000) and Le in e al. (2002)-, and he gene al esul is ha i does no exis . Resul s o indi idual uni oo es s a e a ailable om he au ho s upon eques . The speci ic es ing p ocedu e adop ed is he Augmen ed Dickey-Fulle (ADF) es wi h he Schwa z in o ma ion c i e ion used o selec he numbe o lags included in he ADF eg essions. By de aul , he maximum numbe o lags allowed in he es s is 12. In he Appendix we also show he summa y s a is ics o in la ion, RPV and skewness. 3 As well known, i he lagged endogenous a iable is no co ela ed wi h he e o e m, he alidi y o he OLS es ima o holds. To p o e ha he e is no co ela ion, we ha e es ima ed he model wi h OLS and e i ied ha he e is no au oco ela ion in he esiduals. Page 5 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 6 exis ence o a end can be he key o his esul . Finally, acco ding o he coe icien s on lagged in la ion, i is clea ha in la ion shows pe sis ence. In conclusion, hese esul s seem o con i m he p edic ions o Ball and Mankiw’s model ega ding he ele ance o skewness, and show ha RPV is mo e sensi i e o changes in he in la ion egime ( he wo sample pe iods in ou analysis) han skewness. 4. Panel da a analysis In his sec ion, we pe o m panel da a analysis in o de o con ol o he possibili y ha egional in la ion may be a ec ed by common ac o s, which lead o s ong co ela ion ac oss egional in la ion a es. In o de o implemen i , we a end o he ollowing es ima ion: 17...1 321,1 = + + + + = −jRPVS j j j jjj ε β β π β α π [2] whe e αj is a ixed e ec o each egion. As i can be seen om equa ion (2), lagged in la ion is co ela ed wi h he ixed e ec s. The e o e, wi hin es ima o s will be biased and inconsis en . This p oblem canno be a oided es ima ing he model in i s di e ences, because al hough he ixed e ec is wiped ou , he i s -di e enced a iables a e co ela ed wi h he andom componen o he e o e m. The deg ees o inconsis ency and bias depend on T; only i T∞ he wi hin es ima o is unbiased and consis en .4 In o he wo ds, o a ypical panel whe e N is la ge in ela ion o T (T is usually ixed), and whe e he enla gemen o he sample always e e s o N and no o T, ins umen al a iable es ima ion is equi ed in o de o ge consis en and unbiased es ima o s. Howe e , his is no ou case because N ( egions) is ixed, T is e y la ge in ela ion o N, and he enla gemen o he sample can be e e ed only o T. Despi e he discussion abou he numbe o pe iods equi ed o ge an unbiased and consis en wi hin es ima o would dese e a lo o a en ion, we ha e conside ed ha he ea u es o ou sample allow us o use wi hin es ima o s. 4 See Bal agi (1995, p. 126). Page 6 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 7 Now, we es ima e (2) o he wo subpe iods5 and he o al pe iod –see Tables 5, 6 and 7, i s column- and we pe o m a ixed e ec es 6 o he null hypo hesis α j = α , o all j = 1…17. The es s a is ic is dis ibu ed unde he null hypo hesis as a F16,1169 and i s alue is 1.53 o he 1993-1998 pe iod, as a F16,1390 and i s alue is 1.15 o he 1999-2005 pe iod, and as a F16,2597 and i s alue is 1.04 o he o al pe iod. The e o e, he null hypo hesis ha α j a e equal canno be ejec ed in any case, so we es ima e (3) –see Tables 5, 6 and 7, second column–: 17...1 321,1 = + + + + = −jRPVS j j j jj ε β β π β α π [3] Finally, he ins umen al a iable es ima ion sugges ed by Ande son and Hsiao (1981) is applied –see Tables 5, 6 and 7, hi d column. We es ima e he model in i s di e ences, in o de o ge id o he hypo he ical indi idual e ec s: )()()()( 1,1,31,22,1,11, −−−−−− −+−+−+−=− jj jj jj j j jj RPVRPVSS εεββππβππ [4] As (πj, -1- πj, -2) is co ela ed wi h he new e o e m, we un an ins umen al a iable es ima ion using he in la ion a iable in le els πj, -2 as ins umen ; o he es o he a iables we do no de ine any ins umen s. [Table5] [Table 6] [Table 7] As i can be obse ed, he e a e no ema kable changes wi h espec o skewness o he h ee me hods o es ima ion epo ed, and i s coe icien seems o be s able ac oss pe iods. Bu his does no hold o RPV and he cons an e m in he OLS es ima ion.7 These esul s lead us o in oduce in he es ima ion o he o al pe iod bo h a dummy a iable (D93-98) and a slope dummy (D93-98*RPVj, ) o he 1993-1998 pe iod, in o de o cap u e he change in he cons an and in he coe icien o RPV espec i ely. Mo eo e , we ha e checked ha a slope dummy o skewness is no 5 In o de o ein o ce he alidi y o he di ision in he sample pe iod ha we use in ou analysis we ha e implemen ed a Chow es . The c i ical alue o his es is 3.02 a 1% ( he F s a is ic is 27.49) so we ejec he null hypo hesis o lack o a b eak in 1998:12. 6 See Bal agi (1995, p. 12). Page 7 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 8 signi ican . We ha e un he eg ession wi h ixed e ec s o he whole pe iod, and again he Hausman es leads us o ejec he ixed e ec s, so inally we p esen he esul s o he OLS es ima ion in Table 8: [Table 8] Summa ising, ou esul s show a homogeneous beha iou bo h ac oss egions and pe iods ega ding skewness, which can be e ealing he ulne abili y o he Spanish economy o supply shocks. As a as RPV is conce ned, he p edic ions o Ball and Mankiw (1995) o no end in la ion a e con i med, gi en ha i is no signi ican o he 1999-2005 pe iod in any egion. This a iable appea s o be hea ily a ec ed by he beha iou o he in la ion a e. 5. In oduc ion o con ol a iables As i was men ioned in he in oduc ion, in his sec ion we include se e al con ol a iables. The idea embedded in he inclusion o hese a iables is wo old: i) o check he obus ness o he a o emen ioned ela ion be ween mean in la ion on he one hand and skewness and RPV on he o he –Ball and Mankiw’s app oach–; ii) o ge some p elimina y empi ical e idence on he ele ance o di e en mac oeconomic ela ions o he Spanish economy. Al hough we ha e in oduced many con ol a iables, we would ha e liked o include e en a highe numbe , bu he mon hly equency imposed an impo an sho coming ( hink o ins ance in a iables ela ed o iscal policy). Wi h he excep ion o he egional unemploymen a e, hese a iables a e p o ided a a na ional le el, because hey a e no a ailable, homogeneously, a a egional le el. The da a sou ces o ou con ol a iables a e he ollowing8: i) Unemploymen a es: Ins i u o Nacional de Empleo (INEM). ii) Indus ial p oduc ion index: INE (Base yea 2000). iii) Gene al e ail ade index and shopping mall e ail ade index: INE. i ) In e es a e: Bank o Spain. 7 Resul s o he cons an e m and end a e no included in he ables. They a e a ailable om he au ho s upon eques . 8 A mo e de ailed in o ma ion abou hese a iables and da a sou ces is a ailable om he au ho s upon eques . Page 8 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 15 Quilis, E.M. (1999) No a sob e la elación cíclica en e los índices de p ecios al consumo (IPC) e indus iales (IPRI), Bole ín T imes al de Coyun u a, 73, 141-157. Ra ai, A. (2004) In la ion and ela i e p ice asymme y, Eu opean Cen al Bank, Wo king Pape Se ies, n. 301. Acknowledgmen s The au ho s would like o hank wo anonymous e e ees, Ca los Dabús, Diego Rome o-A ila, semina pa icipan s a he Pablo de Ola ide Uni e si y and Cen o de Es udios Andaluces, and pa icipan s a he EEFS 5 h Annual Con e ence (C e e, G eece, 2006) and ERSA 46 h Cong ess (Volos, G eece, 2006) o aluable commen s and sugges ions. The au ho s also acknowledge inancial suppo om Jun a de Andalucía: Cen o de Es udios Andaluces (ECO 17-2004 and ECOD1.05/033) and CICE (P oyec o de Excelencia 01252). The usual disclaime applies. Page 15 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 16 Table 1: Empi ical e idence Au ho s Coun y and pe iod Da a In la ion egimes Main esul s *Ball and Mankiw (1995) US 1949-1989 Annual da a P oduce p ice Π a ound 3% Posi i e in la ion/skewness associa ion s onge han in la ion/RPV Lou enco and G uen (1995) Aus alia 1970-1992 Qua e ly da a Consume and p oduce p ice L.I.P.: Π < 4-5% H.I.P.: Π > 4-5% The in la ion/skewness ela ion s onge han in la ion/RPV in L.I.P. The opposi e is ue o H.I.P. *Amano and Macklem (1997) Canada 1962-1994 Annual and qua e ly da a P oduce p ice Low and s able in la ion Posi i e in la ion/skewness associa ion and weak in la ion/RPV associa ion Hall and Ya es (1998) UK 1975-1996 Mon hly da a Re ail and p oduce p ice Changing in la ion a e: Π > 12% in mid- se en ies, nega i e a es in 1986-87 and ea ly nine ies Fo he whole pe iod: in la ion/skewness associa ion is weake han in la ion/RPV Auc emanne e al. (2002) Belgium 1976-2000 Mon hly da a Consume p ice H.I.P.: 1976-87, Π = 5,3% L.I.P.: 1987-00, Π = 2,8% Posi i e in la ion/skewness and in la ion/RPV associa ions independen ly o mean in la ion *Döpke and Pie dzioch (2003) Ge many 1969-2000 Annual da a Consume and p oduce p ice Changing in la ion a e Simila posi i e in la ion/skewness and in la ion/RPV associa ions Assa son (2004) Sweden 1980-2002 Mon hly and qua e ly da a Consume p ice H.I.P.:1980-89 L.I.P.: 1990-02 RPV and skewness a e mo e impo an in explaining in la ion in he L.I.P. han in he H.I.P. Ca aballo and Usabiaga (2004a) Spanish egions 1994-2001 Mon hly da a Consume p ice Π < 5% Posi i e in la ion/skewness associa ion is s onge han he in la ion/RPV one Ca aballo and Usabiaga (2004b) Spain 1993-2001 Mon hly da a Consume and P oduce p ice Π < 5% Skewness is signi ican while RPV is no signi ican Ra ai (2004) Hunga y 1992-1997 Mon hly da a Consume p ice Π: 15%-30% Posi i e in la ion/skewness associa ion Ca aballo and Dabús (2005) Spain 1975-2002 A gen ina 1960-1989 Mon hly da a Spain: P oduce p ice A gen ina: Wholesale p ice Spain: H.I.P.: 1975-85, Π: 14% L.I.P.: 1986-01, Π: 2,2% A gen ina: L.I.P.: 1960-75, Π: 23%, H.I.P.: 1976-01, Π: 162% Fo bo h coun ies: H.I.P.: RPV is signi ican while skewness is no signi ican . L.I.P.: Skewness is signi ican and RPV only o A gen ina. Π e e s o he mean annual in la ion a e, L.I.P. o low in la ion pe iod and H.I.P. o high in la ion pe iod. The as e isk implies ha he wo k does no ake in o accoun he e ec s o in la ion egimes on he esul s. Page 16 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 17 Table 2: Regional analysis (1993-1998) Region Cons an π ππ π j, -1 Sj, RPVj, T end Adjus ed R2 Andalucía 0.15 (0.00) 0.47 (0.00) 0.03 (0.00) 0.05 (0.02) -0.002 (0.00) 0.71 A agón 0.29 (0.00) 0.40 (0.00) 0.02 (0.00) -0.01 (0.48) -0.003 (0.00) 0.75 As u ias 0.14 (0.35) 0.50 (0.00) 0.01 (0.00) 0.04 (0.57) -0.002 (0.00) 0.75 Balea es 0.19 (0.00) 0.26 (0.02) 0.004 (0.02) 0.08 (0.01) -0.002 (0.00) 0.83 Cana ias 0.32 (0.00) 0.01 (0.83) 0.04 (0.00) 0.04 (0.09) -0.003 (0.00) 0.51 Can ab ia 0.50 (0.00) -0.33 (0.00) 0.00 (0.95) 0.02 (0.35) -0.004 (0.00) 0.47 Ca aluña 0.30 (0.00) 0.27 (0.01) 0.01 (0.00) 0.007 (0.86) -0.003 (0.00) 0.60 Cas illa-León 0.30 (0.00) 0.19 (0.08) 0.005 (0.01) 0.02 (0.44) -0.004 (0.00) 0.64 Cas illa-La Mancha -0.27 (0.07) 0.43 (0.00) 0.01 (0.00) 0.27 (0.00) -0.000 (0.59) 0.76 Ex emadu a 0.08 (0.24) 0.44 (0.00) 0.03 (0.00) 0.09 (0.00) -0.002 (0.00) 0.77 Galicia 0.28 (0.00) 0.36 (0.00) 0.01 (0.00) 0.00 (0.84) -0.003 (0.00) 0.78 Mad id 0.28 (0.00) 0.25 (0.02) 0.01 (0.00) 0.01 (0.70) -0.003 (0.00) 0.61 Mu cia 0.23 (0.00) -0.37 (0.00) 0.01 (0.00) 0.18 (0.00) -0.004 (0.00) 0.63 Na a a 0.53 (0.01) 0.23 (0.04) 0.007 (0.08) -0.06 (0.56) -0.005 (0.00) 0.63 País Vasco 0.40 (0.00) -0.18 (0.14) 0.00 (0.21) 0.07 (0.05) -0.004 (0.00) 0.61 La Rioja 0.26 (0.17) -0.15 (0.22) 0.00 (0.87) 0.12 (0.17) -0.003 (0.00) 0.35 Valencia 0.05 (0.52) 0.35 (0.00) 0.03 (0.01) 0.12 (0.01) -0.001 (0.00) 0.60 Page 17 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 18 Table 3: Regional analysis (1999-2005) Region Cons an π ππ π j, -1 Sj, RPVj, Adjus ed R2 Andalucía 0.14 (0.00) 0.39 (0.00) 0.03 (0.00) 0.003 (0.89) 0.26 A agón 0.08 (0.28) 0.38 (0.00) 0.02 (0.00) 0.04 (0.30) 0.23 As u ias 0.11 (0.07) 0.42 (0.00) 0.01 (0.00) 0.01 (0.66) 0.24 Balea es 0.17 (0.00) 0.56 (0.00) 0.00 (0.00) -0.04 (0.06) 0.47 Cana ias 0.28 (0.00) 0.26 (0.00) 0.02 (0.00) -0.10 (0.06) 0.16 Can ab ia 0.35 (0.00) -0.52 (0.00) 0.01 (0.19) 0.01 (0.46) 0.24 Ca aluña 0.20 (0.00) 0.21 (0.04) 0.01 (0.00) 0.01 (0.75) 0.09 Cas illa-León 0.12 (0.09) 0.37 (0.00) 0.009 (0.02) 0.02 (0.54) 0.18 Cas illa-La Mancha 0.09 (0.40) 0.37 (0.00) 0.02 (0.00) 0.03 (0.58) 0.22 Ex emadu a 0.05 (0.29) 0.42 (0.00) 0.03 (0.00) 0.03 (0.14) 0.27 Galicia 0.12 (0.02) 0.40 (0.00) 0.009 (0.02) 0.01 (0.50) 0.20 Mad id 0.14 (0.08) 0.21 (0.03) 0.01 (0.00) 0.03 (0.48) 0.18 Mu cia 0.25 (0.00) -0.02 (0.78) 0.01 (0.00) 0.02 (0.44) 0.10 Na a a 0.17 (0.03) 0.20 (0.05) 0.01 (0.00) 0.02 (0.60) 0.09 País Vasco 0.27 (0.00) -0.01 (0.90) 0.006 (0.02) -0.001 (0.92) 0.02 La Rioja 0.31 (0.00) 0.01 (0.88) 0.009 (0.04) -0.001 (0.44) 0.02 Valencia 0.14 (0.03) 0.44 (0.00) 0.02 (0.16) 0.001 (0.97) 0.20 Page 18 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 19 Table 4: Regional analysis (1993-2005) Region Cons an π ππ π j, -1 Sj, RPVj, T end (93-98) Adjus ed R2 Andalucía 0.11 (0.00) 0.52 (0.00) 0.03 (0.00) 0.03 (0.04) -0.001 (0.02) 0.47 A agón 0.10 (0.03) 0.46 (0.00) 0.01 (0.00) 0.04 (0.06) -0.001 (0.01) 0.43 As u ias 0.12 (0.01) 0.49 (0.00) 0.01 (0.00) 0.03 (0.09) -0.001 (0.00) 0.46 Balea es 0.07 (0.35) 0.71 (0.00) 0.005 (0.14) 0.02 (0.60) -0.000 (0.34) 0.70 Cana ias 0.30 (0.00) 0.11 (0.02) 0.03 (0.00) 0.01 (0.35) -0.002 (0.00) 0.36 Can ab ia 0.43 (0.00) -0.33 (0.00) 0.004 (0.53) 0.04 (0.06) -0.001 (0.00) 0.22 Ca aluña 0.17 (0.00) 0.34 (0.00) 0.01 (0.00) 0.05 (0.06) -0.001 (0.00) 0.28 Cas illa-León 0.10 (0.06) 0.45 (0.00) 0.005 (0.01) 0.05 (0.06) -0.001 (0.02) 0.35 Cas illa-La Mancha -0.01 (0.77) 0.44 (0.00) 0.01 (0.00) 0.13 (0.00) -0.001 (0.01) 0.47 Ex emadu a 0.04 (0.39) 0.52 (0.00) 0.03 (0.00) 0.07 (0.00) -0.001 (0.01) 0.56 Galicia 0.14 (0.00) 0.48 (0.00) 0.01 (0.00) 0.03 (0.09) -0.001 (0.01) 0.46 Mad id 0.08 (0.15) 0.33 (0.00) 0.01 (0.00) 0.09 (0.00) -0.001 (0.01) 0.32 Mu cia 0.22 (0.00) -0.01 (0.88) 0.01 (0.00) 0.08 (0.00) -0.001 (0.00) 0.26 Na a a 0.20 (0.00) 0.29 (0.00) 0.01 (0.00) 0.07 (0.01) -0.002 (0.00) 0.34 País Vasco 0.27 (0.00) 0.12 (0.30) 0.004 (0.15) 0.04 (0.05) -0.001 (0.00) 0.32 La Rioja 0.32 (0.00) 0.05 (0.52) 0.002 (0.43) 0.02 (0.17) -0.001 (0.00) 0.12 Valencia 0.08 (0.07) 0.48 (0.00) 0.03 (0.01) 0.05 (0.05) -0.000 (0.07) 0.37 Page 19 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 20 Table 5: Panel da a analysis (1993-1998), wi h nega i e end Va iable Fixed E ec OLS Ande son-Hsiao π j, -1 0.12 (0.00) 0.25 (0.00) 0.12 (0.00) Sj, 0.01 (0.00) 0.01 (0.00) 0.01 (0.00) RPVj, 0.05 (0.00) 0.04 (0.00) 0.006 (0.03) Adjus ed R2 0.58 0.57 - Table 6: Panel da a analysis (1999-2005) Va iable Fixed E ec OLS Ande son-Hsiao π j, -1 0.28 (0.00) 0.29 (0.00) 0.57 (0.00) Sj, 0.01 (0.00) 0.01 (0.00) 0.006 (0.00) RPVj, 0.01 (0.25) 0.01 (0.05) 0.00 (0.7) Adjus ed R2 0.17 0.16 - Table 7: Panel da a analysis (1993-2005), wi h nega i e end (1993-1998) Va iable Fixed E ec OLS Ande son-Hsiao π j, -1 0.35 (0.00) 0.36 (0.00) 0.39 (0.00) Sj, 0.01 (0.00) 0.01 (0.00) 0.01 (0.00) RPVj, 0.04 (0.00) 0.04 (0.00) 0.002 (0.23) Adjus ed R2 0.32 0.34 - Table 8: Panel da a analysis wi h dummies (1993-2005). OLS Cons an πj, -1 Sj, RPVj, D93-98*RPVj, D93-98 T end (93-98) Adjus ed R2 0.38 (0.00) 0.28 (0.00) 0.01 (0.00) 0.01 (0.02) 0.02 (0.01) -0.14 (0.00) -0.003 (0.00) 0.38 Page 20 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 21 Table 9: In oduc ion o con ol a iables: unemploymen , indus ial p oduc ion, e ail ade, in e es a es, oil p ices and indus ial p ice index Cons an 0.38 (0.00) 0.38 (0.00) 0.39 (0.00) 0.38 (0.00) 0.37 (0.00) 0.44 (0.00) 0.35 (0.00) 0.35 (0.00) πj, -1 0.28 (0.00) 0.28 (0.00) 0.27 (0.00) 0.27 (0.00) 0.27 (0.00) 0.27 (0.00) 0.31 (0.00) 0.26 (0.00) Sj, 0.01 (0.00) 0.01 (0.00) 0.01 (0.00) 0.01 (0.00) 0.01 (0.00) 0.01 (0.00) 0.01 (0.00) 0.01 (0.00) RPVj, 0.01 (0.02) 0.01 (0.02) 0.01 (0.02) 0.01 (0.02) 0.01 (0.00) 0.01 (0.04) 0.01 (0.03) 0.01 (0.09) D93-98*RPVj, 0.02 (0.01) 0.02 (0.01) 0.02 (0.02) 0.02 (0.00) 0.02 (0.06) 0.01 (0.17) 0.02 (0.01) 0.02 (0.01) D93-98 -0.14 (0.00) -0.14 (0.00) -0.14 (0.00) -0.14 (0.00) -0.12 (0.00) -0.14 (0.00) -0.12 (0.00) -0.12 (0.00) T end (93-98) -0.003 (0.00) -0.003 (0.00) -0.003 (0.00) -0.003 (0.00) -0.003 (0.00) -0.004 (0.00) -0.002 (0.00) -0.002 (0.00) Spanish cyclical unemploymen -0.006 (0.50) Regional cyclical unemploymen -0.002 (0.69) Cyclical indus ial p oduc ion index 0.002 (0.01) Cyclical gene al e ail ade index 0.01 (0.00) Cyclical shopping mall e ail ade index 0.008 (0.00) Lagged ( -15) change in in e es a es -0.11 (0.02) Change in oil p ices 0.30 (0.00) Lagged ( -3) change in indus ial p ice index 0.06 (0.00) Adjus ed R2 0.38 0.38 0.39 0.39 0.39 0.34 0.42 0.39 Page 21 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Fo Pee Re iew 22 Appendix Table A1: Panel da a uni oo analysis (1993-1998, wi h end) and summa y s a is ics Va iable Le in, Lin and Chu (2002) B ei ung (2000) S a is ic P ob. S a is ic P ob. Sj, -30.62 0.00 -17.84 0.00 RPVj, -14.29 0.00 1.58 0.00 π j, -12.18 0.00 -3.70 0.00 Mean Max. Min. Sj, 0.58 12.98 -9.84 RPVj, 1.46 3.56 0.52 π j, 0.28 0.81 -0.22 Table A2: Panel da a uni oo analysis (1999-2005) and summa y s a is ics Va iable Le in, Lin and Chu (2002) B ei ung (2000) S a is ic P ob. S a is ic P ob. Sj, -25.50 0.00 -13.98 0.00 RPVj, -4.29 0.00 -2.82 0.00 π j, -27.57 0.00 -18.19 0.00 Mean Max. Min. Sj, 0.43 10.48 -12.48 RPVj, 1.62 2.88 0.53 π j, 0.26 0.74 -0.27 Table A3: Panel da a uni oo analysis (1993-2005, wi h end) and summa y s a is ics Va iable Le in, Lin and Chu (2002) B ei ung (2000) S a is ic P ob. S a is ic P ob. Sj, -48.38 0.00 -26.92 0.00 RPVj, -4.20 0.00 0.42 0.66 π j, -24.53 0.00 -5.86 0.00 Mean Max. Min. Sj, 0.50 12.90 -12.21 RPVj, 1.55 3.56 0.52 π j, 0.27 0.81 -0.27 Page 22 o 22 Edi o ial O ice, Dep o Economics, Wa wick Uni e si y, Co en y CV4 7AL, UK Submi ed Manusc ip 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60