Relative Prices and Sectoral Productivity

Relative Prices and Sectoral Productivity Diego Restuccia University of Toronto and NBER University of Oslo August 4-8, 27 Restuccia Macro Growth and Development University of Oslo / 37

Overview Relative prices and sectoral productivity Basic facts Heterogeneity in services Structural transformation within services Productivity implications Restuccia Macro Growth and Development University of Oslo 2 / 37

Relative Prices and Sectoral Productivity Based on Duarte and Restuccia (26): https://ideas.repec.org/p/tor/tecipa/tecipa-555.html Aggregate labor productivity differences across countries are large Sectoral analysis of labor productivity differences is important to understand aggregate outcomes No systematic real sectoral output data across countries Standard approach uses sectoral price and expenditure data (the inputs of PWT) to infer sectoral productivity Restuccia Macro Growth and Development University of Oslo 3 / 37

Motivation Well-known fact: relative price of services rises with development Standard interpretation: cross-country differences in productivity are larger in manufacturing than in services, e.g.: Balassa-Samuelson, Kravis, Heston, and Summers (983), Hsieh and Klenow (27), Herrendorf and Valentinyi (22) These productivity implications are important as the service sector becomes large with structural change Restuccia Macro Growth and Development University of Oslo 4 / 37

Limitations of Standard Approach Heterogeneity within services; in particular, not all services have an increasing relative price The service sector happens to contain some of the economy s most progressive activities as well as its most stagnant. (Baumol et al, AER, 985) Expenditure differs from output... the price parities computed from these surveys [ICP] are not satisfactory for making sectoral productivity comparisons. (Heston and Summers, AER, 996) Restuccia Macro Growth and Development University of Oslo 5 / 37

Plan Address these two limitations, by disaggregating services between those with a falling relative price (non-traditional) and those with an increasing relative price (traditional) developing a framework with an input-output structure that maps relative expenditure prices to value added sectoral productivity Restuccia Macro Growth and Development University of Oslo 6 / 37

Data International Comparisons Program (ICP) Disaggregated data for 25 prices nominal expenditures real expenditures (real = when measured at common international prices) 29 categories / 3 countries World Input-Output Database (WIOD) Harmonized input-output tables for 4 countries, 35 by 35 industries. Annual tables from 995 to 29, at current prices. Restuccia Macro Growth and Development University of Oslo 7 / 37

Facts The price of services relative to that of GDP increases with income The (nominal) share of services in GDP increases with income The real share of services does not vary systematically with income (real = when measured at common international prices) Restuccia Macro Growth and Development University of Oslo 8 / 37

Relative Price of Services Relative Price of Services.8.6.4.2 ZAR CHE JOR BEL DNK USA FRA FIN DEU HKGKWT NOR ITA LBN PRT SAU GBR AUS SWE CAN NLD IRL OMN ISR ESP JPN AUT CHL SVN KOR GRC BRA MEX NZL SGP NPL URY BGD IDN MAR PER PHL IRQCHNECUZAFMYS ARG TWN COL POL HUN MWI CIV IND BIH ESTCZE KEN SEN NAM LBR PRY SYR THA LVA TZA UGA ZMB YEM TUN MKD VEN ROM SVK MUS GMB BEN AGO BOL SRB NER TGO BGR ETH BFA LTU CMR LKA GNB MDG ALB IRN EGY KAZ RUS SLE LSOVNM MRT PAK BWA RWA SDN SWZ KHM MNG GEOUKR BLR MLI GAB BDI MOZ COG CAF GIN GHA LAO MDA GNQ ZWE NGA KGZ ARM AZE TCD TJK Correlation:.67 5 6 7 8 9 2 GDP per capita (PPP adjusted, log) The relative price of services increases with income Restuccia Macro Growth and Development University of Oslo 9 / 37

Share of Services Share of Services (domestic prices).6.5.4.3.2. ZAR ISR USA GMB ZMB GBR SWE BRA FRA NLD GRC AUS AUT DNK CHE MWI PRT BEL BOL ZAF NZL ESP FINCAN IRQ ITAJPN URY LBN TWN DEU KEN JOR COL BGR MEX NAM POL HKG CZE BIH HUN IRL CHL KOR MAR PER TUN MKD ROM MUS SVK SVN NOR MYS ARG LBR UGA LSO TURKAZLVA EST SGP SRB GNB ECU OMN SLE VEN BFA ALB GEO EGY IRN TGO PHL THA CIV KWT RWA PRY SYR UKR BDI BEN BLR LTU SAU MDG MLI YEM MDA SWZ RUS NER KHM IND VNM SEN IDN CHN BWA ETH KGZ CMR COG LKA NPL PAK ZWE MRT LAO GAB GIN AGO MNG MOZ BGD GHA SDN TCD ARM CAF TJK NGA AZE GNQ TZA Correlation:.6 5 6 7 8 9 2 GDP per capita (PPP adjusted, log) The share of services in GDP increases with income Restuccia Macro Growth and Development University of Oslo / 37

Real Share of Services.9.8 GMB TJK Real Share of Services.7.6.5.4.3 ZAR TCD KGZ ZMB BOL MDA ZWE BGR BDI MWI KEN LSO GNBSLE GEO KAZ MLI NAM UKR BRA IRQ ALB ARM COLMKD ROM BIH BLR MUS CZE ISR RWA EGY TUN ZAF POL SVK TWN IRN LBR HUN GIN BFA KHM SRB COG NZL LAO GRC SWZ TUR URY UGA GBR AZE MEX LVA SWE AUT ARG BWA EST NLD USA TGO MDG VNM LTU ESP FRA JPN AUS BEN MAR PER VEN RUS MYS CAN GHA GAB LBN PRT ITA CHL DEU BEL DNK ETH NERMOZ NGA PAK PRY THA CHE CAF CIV JOR CMR ECU SVN KORFIN MRT YEMNGSYR LKA HKG SGP IRL PHL SENIND GNQ OMN NOR SDN CHN AGO IDN SAU KWT NPL BGD.2. TZA Correlation:.6 5 6 7 8 9 2 GDP per capita (PPP adjusted, log) The real share does not vary systematically with income Restuccia Macro Growth and Development University of Oslo / 37

Heterogeneity in the Service Sector Cross-Country Income Elasticity of Relative Prices Elasticity Standard Error Personal Consumption Expenditures Clothing and Footwear Cleaning and repair of clothing -.2.3 Repair and hire of footwear.8.3 Housing and Water Actual and imputed rentals for housing.2.5 Maintenance and repair of the dwelling -.5.4 Water supply -.2.6 Miscellaneous services relating to the dwelling -.8.4 Furnishings, Household Equipment, and Routine Maintenance of the House Repair of furniture, furnishings and floor coverings -.6.4 Repair of household appliances -.9.3 Domestic services.4.4 Household services -.4.2 Health Medical services.28.4 Dental services.23.4 Paramedical services.4.5 Hospital services.8.3 Transport Maintenance and repair of personal transport equipment -.2.3 Other services in respect of personal transport equipment -.7.3 Passenger transport by railway -..4 Passenger transport by road -.8.3 Passenger transport by air -.59.4 Passenger transport by sea and inland waterway -.28.2 Combined passenger transport -.2.3 Other purchased transport services Restuccia Macro Growth and Development -.5.3 University of Oslo 2 / 37

Heterogeneity in the Service Sector Cross-Country Income Elasticity of Relative Prices Elasticity Standard Error Communication Postal services -.5.4 Telephone and telefax services -.36.3 Recreation and Culture Repair of audio-visual and other equipment -.8.4 Veterinary and other services for pets -.4.2 Recreational and sporting services.7.4 Cultural services -.2.2 Games of chance -.3.2 Package holidays -.6.3 Education.22.4 Restaurants and Hotels Catering services -..2 Accommodation services -.24.3 Miscellaneous Goods and Services Hairdressing salons and personal grooming establishments.24.3 Prostitution.26.4 Social protection -..2 Insurance -.4. Financial intermediation services indirectly measured -.4.2 Other financial services n.e.c. -.4.3 Other services n.e.c. -.3.2 Government.2.3 Production of Health Services.2.4 Education.28.4 Collective Services.8.3 Restuccia Macro Growth and Development University of Oslo 3 / 37

Heterogeneity in the Service Sector We divide services into two categories: traditional and non-traditional services Traditional services includes service categories for which the relative price increases with income (government and, from personal consumption expenditures, rents for housing, health services, and education services) Non-traditional services includes the service categories in personal consumption expenditures for which the relative price declines with income (transportation, communication, financial and related services) Restuccia Macro Growth and Development University of Oslo 4 / 37

Relative Prices Traditional Non-Traditional 4 Relative Price of Traditional Services.8.6.4.2 CHE USA KWT BEL DNK FRA FIN IRL PRT ITA NOR DEU SWE HKG CAN NLD SAU KOR ESPGBR AUS LBN ISR MEX OMN SVN NZL GRCJPN AUT IRN CHL IRQ SGP NPL MARSYR BRA BGD ECUZAF URY ARG TWN IDN PER BIH HUN COL SRB ESTCZE ZMB CHN GMB PHL ROM LVA POL MKDMYS CIV INDYEM NAM LBR TZA SEN TUN THA TUR MUSLTU ETH MDG UGA VEN SVK NER BFA AGO BGR KEN LSO PRY MRT ALB RWA SLE BOL MDA SWZ BWA KAZ BLR MWI GNB LKA RUS TGOBEN PAK MNG CMR GEO EGY MOZ MLI SDNVNM UKR ZAR KHMLAO BDI GHA GAB CAF GIN KGZ ZWE COG AZE ARM NGA TJK GNQ TCD 5 6 7 8 9 2 GDP per capita (PPP adjusted, log) JOR Relative Price of Non Traditional Services 3.5 3 2.5 2.5.5 ZWE GNB SLE TCD ZAR UGA LBR BDI MWIRWA KHM LSO SWZ NAM GMB VNM CAF TGO GAB GIN MLI BFABGD BEN MRT COG TZA NPL GHA KEN SEN NGA CIV CMR PHL MDG AGO MNG NER PRY BOL CHN EGY BRA GNQ ETH SDN IDN IRQ COL CHL ARM PER MYS ECU KAZ JPN DEU HKG IND MAR LBN ARG AUS BEL AUT DNK CHE BIH MKDBGR MUS OMN SGP TUN TUR URY ZAF ZMB YEM ISR ESTCZEESP FRA FIN GBR HUN GRC ITA NLD NOR THAVEN RUSSVKSAU MOZ CAN IRL GEO LVA POLPRT SVN TWN SWE KOR KWT PAK LKA UKR LAO ALB MEX LTU NZL USA ROM SRB BLR SYR AZE KGZ JOR MDA TJK 5 6 7 8 9 2 GDP per capita (PPP adjusted, log) BWA IRN Restuccia Macro Growth and Development University of Oslo 5 / 37

Real Share of Non-traditional Services Real Share of Non Traditional Services (in Total Services).5.4.3.2. IRN IRL MEX ESP AUT HKG NLD KOR PRT GBR DNK SGP CHE NOR ITASWE FRA FIN DEU BEL USA MYS POL GRC CAN LKA KWT THA SVN PER VENCHL NZL MDA ARG ECU HUN OMN IND AUS SVK TWN JPN URY CHN CZE IDN EST ALB JOR LTU PAK BGR ISR CIV MAR TUN BOL RUS LBN BRA SRB PHL ZMB PRY LVA YEM TJK COL UKRROM SAU SYR TUR KEN GEO MKD ZAF CMR BIH BLR MUS MWI SDN BEN KGZ GNQ KAZ NER TGO SEN AGO COG AZE EGY NPL BGD ETH TZA VNM GAB KHM NGA LAO BFA ARM CAF MLI MNG NAM MOZ BDI GIN UGA GHA LBR GNB TCD MDG MRT IRQSWZ ZAR ZWERWA SLE LSO BWA GMB 5 6 7 8 9 2 GDP per capita (PPP adjusted, log) Real share of non-trad. services increases with income Restuccia Macro Growth and Development University of Oslo 6 / 37

U.S. Evidence: Share of Non-traditional Services 45 real nominal Share of Non traditional Services in Total Services 4 35 3 25 2 95 96 97 98 99 2 2 Restuccia Macro Growth and Development University of Oslo 7 / 37

U.S. Evidence: Relative Price of Non-traditional to Traditional Services.6 Relative Price of Non traditional to Traditional Services.5.4.3.2..9 95 96 97 98 99 2 2 Restuccia Macro Growth and Development University of Oslo 8 / 37

Development Accounting Basic Framework Minimal structure Key assumptions: competitive markets linear technology in labor perfect labor mobility across sectors Restuccia Macro Growth and Development University of Oslo 9 / 37

Development Accounting Linear technologies in each sector Y i = A i L i, i Competitive markets and perfect labor mobility across sectors max L i p i A i L i wl i, p i A i = w, With i L i = L, it follows: p i Y i = wl; w = i i p iy i L Restuccia Macro Growth and Development University of Oslo 2 / 37

Development Accounting One-to-one mapping between income elasticity of relative price and income elasticity of sectoral productivity A i = w/p p i /p log(a i ) = log(w/p) log(p i /p) Income elasticity of sectoral productivity is one minus income elasticity of relative price Restuccia Macro Growth and Development University of Oslo 2 / 37

Development Accounting Example: income elasticities of relative prices of -.2 and.2, imply income elasticities of sectoral productivity.2 and.8 Real GDP per capita ratio (rich to poor) is a factor of 49.3-fold, hence, implied sectoral productivity ratios a factor of 7.5-fold (exp(.2 log(49.3))) and 22.6-fold (exp(.8 log(49.3))) For the manufacturing sector, income elasticity of relative price is -.7 (income elasticity of manufacturing productivity is.7), hence, productivity ratio of 64.8-fold (exp(.7 log(49.3))) These are stark differences in relative productivities derived from relative price behavior within service categories, even compared to the manufacturing sector Restuccia Macro Growth and Development University of Oslo 22 / 37

Development Accounting Results Relative A i GDPpc m s s T s N D.89.74.94.4.8 D 9.66.5.74.83.6 D 5.3.6.23.3.4 D 2.3.2.5.8.2 D.2..3.5. Ratio D /D 49.3 7.5 27.8 2. 73. Income elasticity.7.86.79.2 Restuccia Macro Growth and Development University of Oslo 23 / 37

Discussion Some evidence on productivity gaps larger in segments of service industries (Retail, Telecom, Banking, Software,...) Baily and Solow (2) Related evidence for the U.S. over time productivity growth fast in (market) services relative to manufacturing (e.g., Triplett and Bosworth, 24) Baumol et al (985) Restuccia Macro Growth and Development University of Oslo 24 / 37

Development Accounting Input-Output Structure Technologies q i = B i l α i i h α i i, i B i - productivity in sector i l i - total labor hours in sector i h i - intermediate composite used in sector i Labor and goods markets are competitive Restuccia Macro Growth and Development University of Oslo 25 / 37

Intermediate Composites Intermediate composite used in sector i ( ) ϕji gji h i = j ϕ ji g ji - use of intermediate input j in the production of i Restuccia Macro Growth and Development University of Oslo 26 / 37

Characterization Representative firm in sector i solves max l i,g ji p iq i p j g ji wl i j () subject to q i and h i First two terms is value added p y i y i p i q i j p j g ji. (2) The first order condition of intermediate input use g ji implies, α i ϕ ji p i q i = p j g ji, (3) Restuccia Macro Growth and Development University of Oslo 27 / 37

Characterization Use optimal demand for intermediate inputs into q i to obtain gross output as a function of only labor and prices, q i = B α i i ( ) αi p α i i α i l i, (4) p hi where p hi = j pϕ ji j Using definition of value added and substituting for output and intermediate inputs, obtain ( ) α i p y i y pi i = p i p hi α i ( α i )α α i α i i TFP in value added A i is given by ( α i )α is total factor productivity in gross output α B i i l i, α i α i i Restuccia Macro Growth and Development University of Oslo 28 / 37 B α i i, where B i

Characterization The first order condition of profit maximization with respect to the labor allocation across sectors is given by p y i A i = w; A i = w p y i where p y i = p i j ( pi p j ) ϕji α i α i. Restuccia Macro Growth and Development University of Oslo 29 / 37

Characterization Hence, sector productivities expressed as a simple function of relative income and relative sector gross output prices, and a term including the input-output structure, A i = w/p p i /p ( ) α i pi p α i [ j ( pj p ) ϕji ] α i α i. (5) log(a i ) = log(w/p) log(p i /p) α i ϕ ji [log(p j /p) log(p i /p)] α i Need to calibrate α i and ϕ ji j (6) Restuccia Macro Growth and Development University of Oslo 3 / 37

Share of Intermediate Inputs USA over Time Manufacturing across Countries.9.9.8.7.6.8.7.6 IND IDN CHN BGR HUNCZE KOR TWN SVK EST FRABEL POL PRT ESP ITA NLD BRA TUR MEX LVA FIN LTU SVN GRC JPN AUS SWE CAN ROM IRL RUS DEU AUT USA GBR DNK.5.4.3.2. Manuf. Trad. Svc. Non trad. Svc. 995 2 25 2 m.5.4.3.2. 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 Traditional Services across Countries Non-Traditional Services across Countries.9.8.7.6.9.8.7.6 st.5.4.3.2. IND IDN CHN RUS GBR BGR EST USA ROM CAN SVK IRL BRA CZE FINSWE TUR LVA JPN LTU SVN AUS KOR NLD HUN DEU AUT DNK POL PRT GRC ESP TWN ITA BEL MEX FRA sn.5.4.3.2. IND IDN CHN CZE EST BEL SVK IRL BGR AUS LVA HUN SWE POL ITA DNK PRT SVN KOR FIN ESPFRA GBR NLD ROM AUT DEU JPN CAN RUSLTU USA TUR BRA GRCTWN MEX 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 Restuccia Macro Growth and Development University of Oslo 3 / 37

Intermediate Inputs in Manufacturing USA over Time Manufacturing across Countries.9.8 Manuf. Trad. Svc. Non trad. Svc..9.8 jm.7.6.5.4 mm.7.6.5.4 IND IDN CHN KOR CZE HUN SVN TWN JPN MEXLVA SVK TUR ESTPRT ESP BRA DEU AUT USA ITA FRABEL CAN POL FIN ROM RUS GBR SWE DNK BGR GRC IRL AUS NLD LTU.3.3.2.2.. 995 2 25 2 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 Traditional Services across Countries Non-Traditional Services across Countries.9.8.7.6.9.8.7.6 st m.5 sn m.5 IRL.4.3.2..4.3.2. IND IDN CHN DNK FRA DEU GBR NLD SWE AUS POL LTU ITA FINBEL AUT TUR GRC EST ESP CAN USA BRA LVA ROM BGRRUS SVKPRT SVN JPN HUN TWN MEX CZE KOR IND IDN CHN BRA BGR LVA LTUEST HUNCZEGRC ESP FRAGBR FIN DEU AUS BEL AUT DNK CAN NLD TURROM MEX RUSPOL SVKPRTSVN KOR ITAJPN TWNSWE IRLUSA 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 Restuccia Macro Growth and Development University of Oslo 32 / 37

Intermediate Inputs in Trad. Services USA over Time Manufacturing across Countries.9.8.7 Manuf. Trad. Svc. Non trad. Svc..9.8.7.6.6 jst.5.4.3.2. mst.5.4.3.2. IND IDN CHN ROM KOR GRC BGR TWN HUN JPN LTU TUR MEX SVNESP POLEST CZE PRT FRA FINAUT AUS BELIRLUSA LVA SVK ITA DEU GBR SWE BRA RUS DNK CAN NLD 995 2 25 2 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 Traditional Services across Countries Non-Traditional Services across Countries.9.8.9.8.7.7 BRA st s T.6.5.4 sn s T.6.5.4 IND IDN CHN TUR MEX DNK USA AUS FINSWE ITA FRA DEU NLD PRT TWN RUS ESP BEL AUT LVA JPN GBR CZE SVN CAN POLEST SVK KOR GRC IRL LTUHUN.3.2 ITA BEL BGR HUN SVKPRT FIN DEU SWE NLD DNK. POL AUS RUS AUT EST SVNESP LTU CZE FRA IDN CHN KOR GRC USA TUR JPN IND BRA MEXLVA TWN ROM 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 CAN IRL GBR.3.2. BGR ROM 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 Restuccia Macro Growth and Development University of Oslo 33 / 37

Intermediate Inputs in Non-Trad. Services USA over Time Manufacturing across Countries jsn.9.8.7.6.5 Manuf. Trad. Svc. Non trad. Svc. msn.9.8.7.6.5.4.4 IND CHN.3.2..3.2. IDN ROM MEX RUSLTU BGR TUR BRA POLHUN EST FIN SVK CZE KOR TWN LVA PRT SVN GRC ESP ITAJPN AUT CAN AUS SWE FRA BEL DEU DNK NLD IRLUSA GBR 995 2 25 2 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 Traditional Services across Countries Non-Traditional Services across Countries.9.9.8.7.6.8.7.6 BRA TUR FRA DEU USA BEL IRL GBR AUS DNK NLD ITA TWN CAN LVA JPN SWE AUT ESTPRT CZEGRC HUN SVK SVN KOR MEX ESP LTU POL FIN st s N.5 sn s N.5 IDN CHN BGR ROM RUS.4.4 IND.3.3.2.2. RUS BGR HUN BRA AUS CAN CHN EST CZE ESPFRAGBR FIN IND IDN BEL AUT MEXLVA LTU PRTSVN KOR GRC DEU DNK ITA JPN SWE NLD TUR IRL ROM POLSVK TWN USA 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25. 7 7.5 8 8.5 9 9.5.5 GDP per capita (PPP adjusted, log) 25 Restuccia Macro Growth and Development University of Oslo 34 / 37

Development Accounting with I-O Linkages A m A st A sn With intermediate inputs: Income elasticity.6.67.4 Ratio D /D 69.3 2.4 78. Without intermediate inputs: Income elasticity.7.79.2 Ratio D /D 7.5 2. 73. Restuccia Macro Growth and Development University of Oslo 35 / 37

Robustness A m A st A sn A o With intermediate inputs: Income elasticity.5.65.4.4 Ratio D /D 67.6.5 77.9 74.2 Without intermediate inputs: Income elasticity.7.79.2. Ratio D /D 7.5 2. 73. 73. Restuccia Macro Growth and Development University of Oslo 36 / 37

Conclusions A non-trivial subset of service categories features a falling relative price with income (non-traditional services) A standard development accounting uncovers the importance of heterogeneity in services for productivity implications Labor productivity differences in non-traditional services are much larger than those in manufacturing The process of development involves a reallocation to non-traditional services Facilitating development requires solving the productivity problem in non-traditional services Doing so may require reforms that are more elaborate than the typical openness-to-trade recipe Restuccia Macro Growth and Development University of Oslo 37 / 37