Why Are Residential PV Prices in Germany So Much Lower Than in the United States?

A Scoping Analysis
Joachim Seel, Galen Barbose, and Ryan Wiser
Lawrence Berkeley National Laboratory September 2012
For offering comments and/or assistance, thanks to Ted James, Alan Goodrich and Kristen Ardani (NREL), (NREL) as well as R h l T ll Rachel Tronstein (U S DOE) Thi analysis was f d d b th S l E t i (U.S. DOE). This l i funded by the Solar Energy Technologies Program, Office of Energy Efficiency and Renewable Energy of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231.

Table of Contents
Motivation, Scope, and Limitations Background and E i ti Lit t B k d d Existing Literature
Learning related to overall PV market size

German Survey Results

Customer acquisition costs Installation labor costs Permitting, interconnection and inspection costs Sales taxes Other soft BoS costs

Secondary Analysis
Project development time E Economies of scale i residential system size i f l in id ti l t i Chinese module market share

Summary Bibliography Bibli h

2

Motivation, Scope, and Limitations p

The installed price of residential PV is significantly lower in Germany than in the U S due primarily to differences in soft costs U.S., soft
But relatively little is known about how/why soft cost components differ

In order to better characterize the nature of these differences, LBNL:

Fielded a survey of German PV installers, adapted from NRELs survey of U.S. y , p y installers, to collect data on residential PV soft costs Comprehensively reviewed public and private consultant data relevant to the cost structure of residential PV in Germany

Focus is the pre-incentive price paid for customer-owned systems

Residential PV in Germany is almost entirely customer-owned; substantial thirdparty ownership in U.S. but pricing sometimes impacted by appraised values

Analysis here i i t d d t b a fi t cut and serves t hi hli ht specific A l i h is intended to be first t d to highlight ifi areas where further research could reveal additional insights
Survey focus was on quantifying differences in specific business process costs Additional research needed to confirm and characterize differences in more detail, as well as to link observed differences to underlying market drivers
3

Germanys 2011 Additions ~4x Greater, and Cumulative Additions More than 5x Greater than in United States Greater,
8000 7000 7485 25000 20000

7408

Annualad dditions[MW W]

6000 5000 4000 3000 2000 1000 0

15000 3794 10000 1950 1900 918 338 2008 473 0 2006 2007 2009 2010 2011 5000

670 108 24 53 32 110 48 110 67 139 94 2000 2001 2002 2003 2004

951

843 149 210

1271

2005

DataSources: US:IEAandGTM/SEIA;Germany:BNetzA ( d l G id S d G /S G (FederalGridAgency) )

CumulativeAddition[M MW]
4

Germanycumulative USAcumulative USAannualadditions Germanyannualadditions

Annual Residential Installations in Germany 2.5x Greater (9.4x (9 4x Greater on per Capita Basis) than in United States
AnnualresidentialPVinstallations
800 700 600 W/capita 500 MW 400 300 200 100 0 2010 2011
DataSources: US:GTM/SEIA;Germany:BNetzA ( d l G id S G /S G (FederalGridAgency) )

761 688

CA USA Germany

268 128 115

297

10.0 9.0 8.0 7.0 6.0 5.0 4.0 40 3.0 2.0 1.0 0.0

9.4 8.5

3.4 34

3.0 0.9 1.0

2010

2011

Cumulative Residential Installations in Germany 3.6x Greater (14x on per Capita Basis) than in United States
CumulativeresidentialPVinstallations
4000 3500 3000 2500 MW 2000 1500 1000 500 0 2010 2011
DataSources: US:GTM/SEIA;Germany:BNetzA ( d l G id S G /S G (FederalGridAgency) )

CA 3420 2731 3 USA Germany W W/capita

45 40 35 30 25 20 15 10 5 0 2010 2011 10 2 13 3 34

42

934 373 637 488

Varied Data Sources Are Available for U.S. and German PV System Pricing
LBNL Tracking the Sun (TTS): Installed prices for ~70% of PV capacity installed in the U.S. from 1998-2011 NREL Cost Modeling Team: Quarterly bottom-up installed price benchmarks based on interviews with installers and modeling EuPD: Project-level price quotes collected through quarterly survey of German installers (since 2008); used for BSW price reports Photon, other consultants: Installed price benchmarks based on interviews with installers or other market research Miscellaneous: Schaeffer et al., 2004, Learning from the Sun; Haas 2004 al 2004 Learning Sun ; Haas, 2004, Progress in Markets for Grid-Connected PV Systems in the Built Environment; Credit Agency for Reconstruction (KfW); IEA National PVPS reports; Langen 2010

Residential PV System Prices Have Often Been Higher in the U S Than in Germany U.S.
MedianInstalledPriceofCustomerOwnedPVSystems10kW*
12 10 $2011/ /W 8 6 4 2 0 4.25 6.91 6.21 6 21 USsystemprices Germansystemprices 6.56 5.88 5.26 modulefactorygateprices CA 3.42 NJ AZ

1.81 1.35 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

*Note:Focusingonsystems10kWservesasaproxyfortheresidentialmarket,astheprojectlevelinstalledpricedataforGerman *N t F i t 10kW f th id ti l k t th j t l l i t ll d i d t f G systemsusedforthisfiguredonotincludehostcustomertype

DataSources: U.S.SystemPricesarederivedfromLBNLsTTSdatasetandareequaltothemedianofcustomerownedsystems10kW installedineachyear.GermanSystemPrices aretheaveragesofindividualpricequotesinEuPD s dataset(20082011)or installed in each year German System Prices are the averages of individual price quotes in EuPDs dataset (2008 2011) or theaverageofpricesreportedbyIEA,Photon,KfW,andSchaeffer(20012007). ModuleFactoryGatePrices aretheaverageofpricesreportedbyIEA,GTM,IRENA,Navigant,andPhoton(annual currencyexchangerateswereusedformodulepricesestimates)
8

As of Q4 2011, the Installed Price Differential Was About $2 8/W $2.8/W

MedianInstalledPriceofCustomerOwnedPVSystems10kW
8.50 7.50 $2011/W W 6.50 5.50 4.50 3.50 2.50 2010 2010 2010 2010 2011 2011 2011 2011 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 InstallationDate I ll i
*Note:Germansystempricesareavailablebasedonthedateofpricequote,ratherthanbyinstallationdate.However,the averagetimelagbetweenpricequoteandinstallationdateismuchshorterinGermanythanintheUS.,asdescribedfurtherwithin thesecondaryanalysis

7.36

7.05

USsystemprices
6.80 6.50 6.44 6.28 6.16 5.90 6.31 5.56 5.11

Germansystemprices CA NJ

4.38

4.32

4.20

4.09

3.67

3.61

3.26

3.08

AZ

DataSources: US:TTS;Germany:EuPD

Installed Prices in the U.S. Are Also Much More Varied Than in Germany
FrequencyDistribution: InstalledPriceof10kWCustomerOwnedSystemsInstalledin2011 Installed Price of 10 kW Customer Owned Systems Installed in 2011
40% 35% 30% 25% 20% 15% 10% 5% 0% 1 1.01.5

Per rcentofsyste ems

USA Germany

Some U.S. systems have reached German prices already Greater variation in the U.S. indicative US of greater market fragmentation across jurisdictions

1 1.52.0

2 2.02.5

2 2.53.0

3 3.03.5

3 3.54.0

4 4.04.5

4 4.55.0

5 5.05.5

5 5.56.0

6 6.06.5

6 6.57.0

7 7.07.5

7 7.58.0

8 8.08.5

8 8.59.0

9 9.09.5

9.510

$2011/W
*Note:Germandatacomefromasurveyofsystempricequotesfromroughly100installersperquarter, andarethusbasedonamuchsmallersamplethantheUSdataandmaynotreflectthefullextentof pricevariabilityintheGermanmarket.

DataSources:US:TTS;Germany:EuPD

>10

10

Learning Curve Analyses of BoS Costs

Question: To what extent are lower BoS costs in Germany potentially due to larger overall market scale and associated learning-induced cost reductions?
Traditional PV learning c r e anal ses often foc s on PV mod les and curve analyses focus modules relate global module production and module prices Some business process costs (e.g., installation labor, customer acquisition) may also be subject to local learning effects We compare the relative impact of local BoS learning in the U.S. and y p y Germany based on implied non-module costs for <10 kW PV systems and cumulative national PV capacity installed BoS progress ratios may help predict future U.S. price reductions that accompany larger market scale

11

Differences in Market Size Alone May Explain Roughly Half of the Price Gap
ImpliedAverageAnnualNonModuleCosts* vs.CumulativeCapacity: vs Cumulative Capacity: CustomerOwnedSystems10kW,20012011
10 9 8 7 6 5 4 3 2 1 0 100 USA Germany

Total non-module costs in 2011 were ~$2.8/W hi h i the U S $2 8/W higher in h U.S. than in Germany But, at the same cumulative capacity th t the U.S. had it that th U S h d installed at the end of 2011 (4 GW), non-module costs for y residential PV in Germany were only $1.3/W less than in the U.S.

$2 2011/W

2011 $1.30 $1 30 $1.50 2011

1000 10000 cumulativenationalPVcapacityinstalled[MW]

*Note:Impliedaverageannualnonmodulecost=averageannualsystem priceminusglobalaveragefactorygatemoduleprice

One might (crudely) infer that the remaining $1.5/W of the total g g gap in 2011 non-module costs may be due simply to the larger base of German experience

DataSources:Seeslide8.

12

Soft-Cost Learning for <10 kW Systems Occurs More Slowly in the U S and Is Less Effective U.S.
Regressionvariable(20012011) Globallevelcumulative installations (allPVsystems,notjust residential) Countrylevelcumulative installations (allPVsystems,notjust residential) Totalsystemprices Non modulecosts Nonmodule costs Totalsystemprices Nonmodulecosts UnitedStates PR:91.7%,R2:0.90 PR:94.2%,R2:0.48 PR: 94 2% R2: 0 48 PR:90.4%,R2:0.92 PR:93.3%,R2:0.48 Germany PR:82.1%,R2:0.92 PR:79.7%,R2:0.95 PR: 79 7% R2: 0 95 PR:86.9%,R2:0.83 PR:84.6%,R2:0.91

* Notes: PR is the Progress Ratio, defined as 2^(slope of line of best fit of log-log plot).

The development of non-module costs is less correlated with market growth in the US than in Germany (52% vs. 9% explained by other factors) The learning rate for non-module costs (proxy for soft costs) is lower in the US than g (p y ) in Germany (7% vs. 15%)
13

Regular FiT Adjustments Pressure German Installers to Reduce Prices

GermanFiT andsystemprices(<10kWsystems)
14 12 $2011/W 10 8 6 4 2 0 J Jan04 Jul04 J Jan05 Jul05 J Jan06 Jul06 J Jan07 Jul07 J Jan08 Jul08 J Jan09 Jul09 J Jan10 Jul10 J Jan11 Jul11 J Jan12 Jul12 J Jan13 1.00 $2011/k kWh 0.80 0.60 0 60 0.40 0.20 0.00

BNEF (2012) indicates the presence of value-based value based pricing in both the US and Germany Following this hypothesis, the iterative reduction of the FiT presses German installers to lower system prices t maintain attractive i to i t i tt ti investments for their customers Similar forces may operate less efficiently in the U.S., yielding higher valuebased prices, even for customer-owned systems d
14

southernFiTNPV($2011/W) Germansystemprice$/W

northernFiTNPV($2011/W) FiT$2011/kWh

DataSources:EuPD (20082011),IEA,KfW,Photon

Hypotheses Explored for Why German and U S Residential PV Prices Differ U.S.
General:
Residential systems are larger in Germany yes US installers develop projects more slowly yes (semi-addressed) US installers have higher net-profit margins, after recovering all overhead expenses likely (semiaddressed)

Installation labor:
US installers need longer for the installation process yes US installers have higher wages yes for installation labor, no for other labor (semiaddressed)

Component costs:
Hardware component costs are lower in Germany possibly true for inverters, but uncertain (semiaddressed) US has a lower share of cheaper Chinese modules no

P Permitting, I t itti Interconnection and I ti d Inspection ti Costs

US installers have higher labor hour requirements for PII yes US has higher permitting and interconnection fees yes

Customer acquisition:
US installers have higher customer acquisition costs yes US installers have lower customer success rates yes US installers have higher marketing and advertising costs yes

Taxes
The US charges higher sales taxes on PV systems than Germany yes

15

Additional Hypotheses Not Explored Here

Overhead costs
US has higher business overhead costs (e.g. insurance costs, material storage costs) i t t i l t t ) German installers have higher sales volume per year, spreading fixed costs over larger denominator and profiting from economies of scale, allowing for volume discounts US installers have higher cost of capital for their own business operations US installers face higher transaction costs associated with arranging financing for customers US has a longer supply chain for PV modules and other hardware

Regulatory issues
US requires each panel and rack component to be b grounded t th DC switchbox l di t d d to the it hb leading to higher material costs and installation labor hours Less onerous requirements for roof mounting structures

Installation timing
US systems are installed more steadily throughout the year, whereas German installations were traditionally concentrated at the end of the year when prices are lower, leading to seemingly lower annual price averages

Profit margins
US has a lower degree of competition among installers, maintaining higher profit margins Value based pricing allows for higher prices in the p g g p US, given better irradiation, high retail rates in some regions, and more generous subsidies

Exchange rate dynamics are more beneficial for German system costs

16

A Small Body of Literature Explores the German-U.S. German U S PV Price Gap

Few have sought to explain the underlying reasons behind the German-U.S. PV price gap or to quantify differences in specific soft costs
Photon 2011a, Photon 2011b, BNEF 2012, Langen 2010, Podlowski 2008, Goodrich et al. 2012 Value-based pricing in the U.S. (e.g., associated with more generous subsidies and/or less competition among installers) Preference for premium products in the U.S. Lower customer-acquisition costs in Germany due to simpler/more certain value proposition (FiT), critical mass of demand, and economies of scale Lower installation labor costs in Germany due to greater experience and economies of scale Lower permitting costs i G L itti t in Germany d t f due to fewer requirements and greater standardization i t d t t d di ti Less onerous electrical requirements and interconnection processes in Germany Deriving estimates for specific business process costs via a survey of 24 German residential installers Using large samples of s stem prices to compare price de elopments and distrib tions system developments distributions Estimating the impact of differences in project development times on reported prices Analyzing residential module market composition

Possible reasons for the price gap that have been postulated:

Our analysis complements that literature by:

Complements NREL cost modeling team efforts that draw on in-depth interviews with installers i t ll

17

SurveyResults Survey Results

18

LBNL Survey of German PV Installers y

Overview of survey approach y pp Sample characterization Survey respondents build-up of installed price respondents build up Individual business process costs (with comparisons to NREL y ) survey of U.S. installers*)
Customer acquisition costs Permitting, interconnection and inspection Installation labor costs

Sales/value-added tax for PV

*Notethataslighttemporalmisalignmentexists,astheNRELsurveyofU.S.installerswasfocusedon2010 g p g , y installations,whereastheLBNLsurveyofGermaninstallerswasfocusedprimarilyon2011installations.

19

Overview of Survey Approach y pp

German survey focuses on standard DOE soft cost categories:
Customer acquisition Permitting, interconnection, inspection g p Installation labor

InstallerSurveySample
Germany 2011 Residential installers Residential systems Residential capacity[kW] 24 2056 17,819 U.S. 2010* 56 6038 34,396

Adapted from NREL survey of U.S. installers to allow comparisons

Average labor hours per system for PII and installation Total annual expenditures on customer acquisition i iti

*SamplesizesshownforU.S.2010referto analysisbyArdani etal.2012 analysis by Ardani et al 2012

Respondents asked about costs of residential systems installed in 2011 Survey instrument, written in German, distributed by email to 300 German residential installers and fielded online via www.photovoltaikstudie.de
20

Raw Sample Characterization p

Most are small-volume installers - Plurality of respondents completed 10-49 residential systems in 2011 - Median value: 26 systems completed in 2011 I t ll Installer-average residential id ti l system sizes (total MW divided by total systems) are relatively large* l * - Half of installers have average residential system sizes >8 kW
NumberofResidentialSystemsInstalled
Numbe erofResponden nts 10 8 6 4 2 0 <9 1049 5099 100199 >200 NumberofResidentialSystemsinstalledin2011 Numberof fRespondents

AverageResidentialSystemSize(kW) A erage Residential S stem Si e (kW)

8 6 4 2 0 <4 48 48 8 12 812 12 16 1216 >16 AverageSizeofResidentialSystemsinstalledin2011

*NotethattworespondentsmayhaveoverreportedMWsinstalled,leadingtoalargecalculatedaveragesystemsize 21 (potentiallyduetomultifamilyhouses)

Total Soft BoS Costs + Profit Represent Roughly $0.62/W or 20% of System Price $0 62/W
ResidentialPVSystemPriceBuildUpReportedbyGermanInstallers (Averages*and25 (Averages* and 25th/75th Percentiles for Systems Installed in 2011 ) PercentilesforSystemsInstalledin2011)
3.5 3.0 2.5 25 3.44

ReportedAverageSystemPriceby24Installers:$3.00/W
0.23 0.33 0.18
Includes installation laborcost, PII,direct customer acquisition

0.15
Includes overhead, costoffailed bids,general advertising

0.28
Excludes overhead costs

3.05 2.70

$ $2011/W

2.0 1.5 1.0 0.5 0.0 module d l 1.82

EuPD 2011 Photon 2011 BNEFQ4 2011

inverter i

other h otherproject othernon h j h hardware cost projectcost

netprofit othersystem fi h price estimates

*Notes:Surveyresultsaresummarizedintermsoftheaverageofresponsesacrosssurveyrespondents,weightedbyeachrespondents reported2011residentialcapacityinstalled.Thischartsummarizesresponsestothesurveyquestionaskinginstallerstoidentifythe t d 2011 id ti l it i t ll d Thi h t i t th ti ki i t ll t id tif th averagepriceofresidentialsystemssoldin2011,andtoallocatethatpriceacrossthecategoriesidentifiedalongthexaxis.

22

Reported Installation Labor Costs and Profit Are Lower than Estimates Reported Elsewhere
Comparison ofSurveyResponsestoOtherEstimatesforResidentialPVinGermany
2.50 2.00 Survey EuPDlow EuPDhigh EuPD high BNEFQ42011

$20 011/W

1.50 1.00 0.50 0 50 0.00 1.82 Module 0.33 Inverter 0.23 Other Hardware 0.04 Installation Labor

PHOTONQ22011smallcommercial

0.29 Other Overhead

0.28 Profit

0.61 TotalSoftBoS

23

Soft Costs for Residential PV in Germany Are ~$2 7/W Lower Than in the U S ~$2.7/W U.S.
Total soft costs for residential PV in Germany, including margin, are just 19% of the implied soft costs for U S residential PV ($0 62/W vs $3 34/W) U.S. ($0.62/W vs. $3.34/W)
7.00 $6.19 6.00 5.00 $2011/ /W 3.34 4.00 3.00 2.00 1.00 0.00 USA2011 Germany2011
24

LBNLTTS:Residentialsystemsofanysize, excluding3rdpartyownedsystems

Implied softBoS +profit (residualofTTSsystem pricesandhardwarecosts)

softBoS+profit otherhardware other hardware inverter module

$3.00 0.47 0.55 1.83

NRELcost modelingH22011

0.62 0.23 0.33 1.82

GTM/SEIA*

*Notes:USmoduleandinverterpricesarebasedonaveragefactorygatepricesforQ42010Q32011asreported byGTM/SEIAwithanadderof10%toaccountforsupplychaincosts.Inverterefficiencyassumedtobe85%.

Labor Rates Are Higher in Germany Than in the U.S. U S for Some Functions but Lower for Others Functions,
FullyburdenedwagesatPVinstallationcompanies[$2011/hr]
USA 70 60 50 40 30 20 10 0 electricianinstallation labor nonelectrician installationlabor systemdesign engineer salesrepresentative administrativelabor 62 48 31 42 47 38 27 35 32 Germansurvey 60 GermanStatisticalAgency* 53 42 35 25 20

The results that follow this slide rely on German wage rates derived from the survey In the above graphic, data from the German statistical agency are also shown for comparison (these data cover all sectors, so are not specific to PV) U.S. labor rates are from RS Means (as used by NREL cost modeling team and as used in NREL BoS survey analysis for the U.S.)
25

Residential Customer Acquisition Costs Average $0.07/W in Germany $0 07/W

Most respondents reported customer acquisition costs <$0.15/W; several smallvolume installers reported somewhat higher costs On average, customer acquisition labor includes 3 hrs/system for sales representative and 2 hrs/system for design engineer
AverageCustomerAcquisition g q CostsAcrossInstallers
$0.08 $0.07 $0.06 $0.05 $0.04 $0.03 $0.02 $0 02 $0.01 $0.00 $.07 0.01 0.02 System Design Marketing and Advertising Other Other Customer Acquisition $2011/W

AverageCustomerAcquisitionCostsforEachInstaller g q
$0.8 $0.7 $0.6 $0.5 $0.4 $0.3 $0.2 $0.1 $0.0 0 200 400 TotalResidentialInstallations(#) 600

$2011/W W

0.04 0 04

Germany2011

Notes:OtherCustomerAcquisitionCostsincludesuchitemsas:salescalls,sitevisits,traveltimetoandfromthesite,contractnegotiation,bid preparation.Marketing&AdvertisingandOtherCustomerAcquisitioncostsarebasedonreportedannualexpenditures,whileSystemDesign costsarebasedonreportedlaborhoursandwagesforsystemdesignengineering. 26

Customer Acquisition Costs in Germany Are $0 6/W Less Than in the U S $0.6/W U.S.
Mean bid success rate is slightly l li htl lower i th US in the (30% in US vs. 40% in Germany) German installers leverage partnerships with equipment manufacturers Langen (2010) points to simpler and more certain value proposition in p p Germany (i.e., FiT), installer learning, and critical mass for word of mouth
AverageCustomerAcquisitionCosts
$1.20 $1.00 $0.80 $201 11/W $0.69 $0.60 $0 60 $0.40 $0.20 0.24 $0.00 USA (2010) 0.11 1.1 1.07 SystemDesign

Nonproject specificMarketing &Advertising Otherproject p j specificCustomer Acquisition Woodlawn Associates2012 Associates 2012 $0.07 0.02 0.04 Germany G (2011)
27

0.34

0.4

Langen2010

Notes:BarchartofUScustomeracquisitioncostsderivesfromNRELsurveyofU.S.installers(Ardani etal.2012).

PII Costs Are Negligible for Residential PV in Germany

Total PII costs of $0.03/W on average Fewer than 10 hours of labor required for all PII activities, and no fee q , Average labor requirement of 5 hrs (confirmed by PV legal survey, lowest for all European countries) Permit requests and incentive application are done online; usually no permit inspection required p q

TotalPIILaborHoursPerRespondent
10 9 8 7 6 5 4 3 2 1 0 0 100 200 300 400 500 ResidentialPVsystemsinstalledin2011(#) 600 $0.04 Totallaborhours/install

AveragePIICosts
0.03$/W 0 03 $/W 0.00 $0.03 $2011/ /W 0.02 $0.02 0.001 0.001 0.01 0 01 PermitFee Incentive Application Completing Interconnection CompletingPermit Inspection SubmittingPermit Package PreparingPermit Package
28

$0.01

$0.00 Germany2011

PII Costs Account for Roughly $0.2/W of the German-U.S. PV Price Gap German U S
hrs/system
25 22.6h 20 3.7 2.9 29 15 4.3 10 3.2 5.2h 0.0 3.0 0.2 0.3 1.7 0.02 0.01 0.03 5 0.03 0.06 $.03/W $ 03/ 0.00 0.02 0.00 0.01

PIIrequirements
SunRun $.28/W $.24/W

$2011/W

0.30

0.25

Incentive Application

Differences due to both PII labor costs and permit fee

PII labor costs are $0.12/W lower in Germany* Remainder of gap ($0.09/W) is associated with permit fee (assuming an average of $ $430 p system in the U.S.) per y )

0.09

Completing Interconnection I t ti 0.20 Completing 0.15 Permit p Inspection Submitting 0.10 PermitPackage Preparing 0.05 PermitPackage 0 05 Permit Package PermitFee
0.00

8.5 85

Langen (2010) estimates PII costs for the US at $.80/W, and Germany at $.10/W SunRun (2011) figure of $.50/W includes sales & marketing costs & variations in building i b ildi requirements i t

0 USA2010 [h/system] [h/ t ]

Germany USA2010 Germany 2011 2011 [$/W] 2011 [$/W] 2011[$/W] [h/system]

*Fullyburdenedlaborratesassumptions:70%designengineerand30%administrativelabor;averaging$41/hrfor Germany(basedonsurveyquestions)vs.$26/hrfortheU.S.(basedonRSMeansdata,perNRELPVcostmodelingteam)29

German Installers Report Surprisingly Low Installation Labor Requirements

All 24 respondents indicated fewer than 15 hrs/system required for installation labor ( l t i i i t ll ti l b (electrician + construction l b ) t ti labor)
Averaged 7.5 hrs/system, equating to less than $0.04/W, based on reported fullyburdened wage rates*

This is much lower than other estimates; warrants further examination

BNEF estimates 24 person-hours for 3kW system; EuPD estimates $0.39-$0.45/W

TotalInstallationLaborpersystem p y
20 hrs/system 15 hrs/syst tem 10 5 0 0 100 200 300 400 500 ResidentialInstallationsin2011(#) 600 8 6 4 2 0

AverageInstallationLaborTimeandCost
electrician 1.71 nonelectrician 0.009 0.04 $2011/ /W
30

0.03 0.02

5.74

0.028

0.01 0

labortime

laborcost

*FullyburdenedGermanlaborrates:electricianwage=$48/hr,nonelectricianwage=$38/hr(averagesofsurvey responses)

Not Surprisingly Then, Installation Labor Costs Are Much Lower Than in the U S U.S.
InstallationLabor
electricianinstallationlabor 80

75h

nonelectricianinstallationlabor 0.7

70 60 h/s system 50 40 30 20 10 0 USA2010 [h/system] 26 BNEF 24h 49

$0.59/W
0.6 EuPD $0.42/W 0.5 0.4 0.3 0.2 $20 011/W

Survey results indicate that, on average, systems are t installed roughly 10 times faster in Germany than in the U.S. (7.5 vs. 75 hours U S (7 5 vs per system) Leading to total installation labor costs that are $0 55/W $0.55/W lower than in the U.S. Other estimates of labor costs for German PV also show savings relative to the U.S., though differential is smaller warrants further investigation
31

0.33

7.5h 7 5h
2 6 Germany 2011 2011 [h/system]

0.26

$0.04/W $0 04/W
0.01 0.03

0.1 01 0

USA2010 [$/W]

Germany 2011[$/W] 2011 [$/W]

Nationwide Sales Tax Exemptions in Germany Further Reduce Soft Costs

Survey respondents confirmed that German residential PV systems are effectively exempt from revenue taxes/ sales taxes/ value added taxes Regular tax rate of 19% can be exempted either via Kleinunternehmer or Vorsteuererstattungs clause In th U S I the U.S., 23 states assess sales t t l tax on residential PV systems, usually 4-8% of system prices, as do many local governments Given the spatial distribution of PV systems, and accounting for sales tax exemptions in some states states, state and local sales taxes added $0.21/W to the price of residential PV in the U S in 2011 U.S.

32

PII, Customer Acquisition, and Installation Labor Total Just $0 14/W for Residential PV in Germany $0.14/W
For residential PV in Germany, PII, customer acquisition, and installation labor l b are estimated t represent 23% of all non-hardware costs ( hi h ti t d to t f ll h d t (which primarily consists of overhead and profit) and 5% of the total system price.
100% 90% 80% 70% $0.28 overheadandother residualsoftcosts salestax($0) permittingfee($0) $0.20 PII customeracquisition andsystemdesign y g installationlabor profit

Hardware: $2.38/W $2 38/W

Non Hardware: $0.62/W $

60% 50% 40% 30% 20% 10% 0% $0.03 $0.07 $0.04 33

Summary of Soft Cost Differences for Residential PV in the U S and Germany U.S.
Comparison of Soft Costs for Residential PV in Germany and the U.S. (customer-owned (customer owned systems)
1.8 1.6 1.4 1.61

USA

Germany

$ $2011/W

1.2 1 0.8 0.6 0.4 0.2 0 customer customer acquistion: acquisition: marketing+ systemdesign advertisement customer acquisition: other 0.34 0.02 0.11 0.01 0.24 0.04 0.15 0.03 0.59 0.48 0.21 0.09 0.00 0.04 installation labor 0.00 salestax profit, overhead,and otherresidual softcosts
Profit 0.28

PIIlaborcost permittingfee

Notes:Profit,overhead,andotherresidualsoftcostsiscalculatedasaresidualtermbasedonthedifferencebetweentotalsoft costsandthe sumoftheindividualbusinessprocesscostsquantifiedthroughtheGermanandU.S.installersurveys.Thisresidualtermincludessuchitems aspropertyrelatedexpenses(rent,utilities,etc.),inventoryrelatedcosts,additionalinsurancesandfees,andgeneraladministrativecosts. 34

Summary of Soft Cost Differences for Residential PV in the U S and Germany U.S.
Breakdown of Cost Differential Between German and U.S. Residential PV (customer-owned (customer owned systems)
7.00 6.00 5.00 5 00 $2011/W 4.00 3.00 2.00 2 00 1.00 0.00 German system module inverter other installation customer hardware labor acquisition and system design PII permitting salestax profit, fee overhead andother residual costs 3.00
Cost difference of $3.19/W

1.13 0.62 0.01 0.22 0.24 0.55 0.12 0.09 0 09 0.21 0 21

Notes: Profit,overhead,andotherresidualsoftcostsiscalculatedasaresidualtermbasedonthedifferencebetweentotalsoft costsandthe sumoftheindividualbusinessprocesscostsquantifiedthroughtheGermanandU.S.installersurveys.Thisresidualtermincludessuchitems aspropertyrelatedexpenses(rent,utilities,etc.),inventoryrelatedcosts,additionalinsurancesandfees,andgeneraladministrativecosts. 35

SecondaryAnalyses

36

Questions Explored through Secondary Data Sources

1. To what extent do shorter project development times in p j p Germany contribute to the apparent price gap (i.e., quicker pass-through of module price declines)? 2. Are residential PV systems larger in Germany, leading to potential price differences due to economies of scale at the system level? 3. Are a larger percentage of German systems comprised of Chinese modules than in the U.S.?
Note:Item2isnotadditivetothedifferencesinspecificbusinessprocesscostspresentedpreviously,butratherhelps toexplainthosedifferences(e.g.largersystemsizesinGermanymightpartlyexplainwhymarketingcosts,onaper Wattbasis,arelower).

37

Longer U.S. Project Development Time Contributes to Apparent Price Gap

MedianPVpricesforsystems10kW
8.50 7.50 6.50 $20 011/W 5.50 4.38 4.50 3.50 3 50 2.50 2010 2010 2010 2010 2011 2011 2011 2011 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 4.32 4.20 4.09 3.67 3.61 3.26 3 08 3.08 7.36 7.05 USAcompletion USAapplication 7.05 6.80 6.80 Germany 6.50 6.44 6.28 6.16 5.90 5.90

Based on TTS data and German survey responses, responses residential projects take 126 days to develop in the U.S. vs. 35 days in Germany When comparing German and U.S. g system prices based on installation date, some of the difference is due to the longer development time in the U.S., i.e., German system pricing is effectively shifted one quarter relative to the U.S. In Q4 2011, this effect contributes ~$0.18/W ($ $ ($3.26 minus $3.08) to the $ ) apparent price gap Larger or smaller impacts in other quarters, depending on speed of price declines
38

6.50 6.44

6.28 6.16

German Residential Systems Are Generally Larger Than U.S. Systems US

S e st but o o SizeDistributionofPVSystems10kWInstalledin2011 Syste s 0 sta ed 0
20% Percent tofSystem ms 15% 10% 5% 0% 01kW 12kW 23kW 34kW 45kW 56kW 67kW 78kW 89kW 910kW USA Germany
USmedian: 4.95kW Germanmedian: 6.8kW

USdatabasedonTTS;GermandatareflectsallgridconnectedPVsystems(infront+behindthemeter) ascollectedbytheFederalGridAgency(Bundesnetzagentur,BNetzA)

39

If the Size Distribution of U.S. Residential Systems Were the Same as in Germany, Median Prices Would Be $0.15/W Lower
7.50 7 50 6.50 $2011/W W 5.50 5 50 4.50 3.50 3 50 2.50 23kW 34kW 45kW 56kW 67kW 78kW 89kW 910kW

MedianPVpricesforU.S.systems10kWin2011
6.94 6.52 6.25 5.97

5.84

5.82

5.71

5.81

Applying the price distribution shown here for U.S. systems to the system size distribution for German systems (shown on the previous slide) yields a median system price that is $0.15/W lower than the actual median price for the 2011 U.S. systems in the TTS data sample ($6.21/W) $
40

Installer Purchase Prices for Chinese Modules Are Lower than for Non-Chinese Modules in Germany Non Chinese
ModulepurchasepriceforGermaninstallers p p
3.00 2.50 2.00 $20 011/W 1.50 1.00 0.50 0.00 Q12010 Q22010 Q32010 Q42010 Q12011 Q22011 Q32011 Q42011 Chinesemodules Datasource:EuPD
41

nonChinesemodules

The Price Gap Is Not Due to Differences in Chinese Module Market Share
Share of module manufacturers by country of headquarters for customer-owned 10kW syste s in 2011 o custo e o ed 0 systems 0
Chinese modules are cheaper, but cheaper but Among customer owned systems 10 kW, the U.S. and Germany had similar shares of Chinese modules* Thus differences in Chinese module market share do not contribute significantly to the German-U S German-U.S. price gap for 10 kW customer-owned systems

USA n=20,761
China+Taiwan 8% 20% 25% USA Japan Germany 22% 24% Restofthe World 53%

Germany n=3,041
27%

6% 5% 9%

Sources:TTS,EuPD *ThirdpartyownedsystemsintheU.S.haveahighershareofChinesemodules(seeBNEF2012,forexample),butforthe purposeofassessingthepricegapinthisanalysis,wefocusspecificallyoncustomerownedsystems.

42

Summary of Findings from Survey of German Installers

Total non-hardware costs for residential PV in Germany are ~$2.70/W lower than in the U S U.S. Customer acquisition costs average just $0.07/W in Germany, or roughly $0.60/W lower than in the U.S. Installation labor I t ll ti l b requirements reportedly average 7 5 h i t t dl 7.5 hours f G for German systems, leading to $0.55/W lower costs than in the U.S. (though these survey data diverge substantially from other estimates) PII processes require 10 hours of labor on average in Germany with no labor, average, Germany, permitting fee, resulting in PII costs roughly $0.20/W lower than in the U.S. German residential systems are exempt from sales/value-added tax, while U.S. systems are subject to an average sales tax of roughly $0.20/W (accounting for sales tax exemptions in many U.S. states) The remaining gap in soft costs between Germany in the U.S. (~$1.15/W) is associated with overhead, profit, and other residual soft costs not captured p p in the categories above
43

Summary of Findings from Secondary Analysis

Shorter project development times in Germany contribute to apparent price gap ( i (e.g., ~$0.2/W effect for Q4 2011 installations) $0 2/W ff t f i t ll ti ) Residential PV systems are larger in Germany (partly due to differences in policy design), benefitting from economies of scale ($0.15/W effect)
Not additive to the differences in soft costs presented previously, but rather helps to explain those differences (e.g. larger system sizes in Germany are partly why marketing costs, on a per Watt basis, are lower)

Market share of Chinese modules is similar for customer-owned residential systems in Germany and U.S., and thus does not contribute to price gap

44

Suggestions for Further Research gg

Initiate a more refined analysis of overhead costs and margins y g among installers Better understand the pricing decision of installers and competition between installers (i e degree of value based pricing ) (i.e., value-based pricing) Clarify installation labor hours in Germany and the U.S. and p potential efficiency g y gains Compare supply-chain margins between the two countries and average prices paid by installers for modules and inverters Assess the role of FIT policies in Germany in stimulating price reductions and potential implications for U.S. solar policy

45

Questions?
Joachim Seel, jseel@lbl.gov Galen Barbose, glbarbose@lbl.gov Ryan Wiser rhwiser@lbl.gov

Download LBNL Electricity Markets & Policy Publications:

http://eetd.lbl.gov/ea/emp/re.html http://eetd lbl gov/ea/emp/re html

Bibliography g p y
A. Tweedie, and E. Dorris. Comparing Germanys and Californias Interconnection Processes for PV Systems. National Renewable Energy Laboratory (NREL), July 2011. Alan Goodrich, Ted James, and Mi h l W dh Al G d i h T d J d Michael Woodhouse. R id ti l C Residential, Commercial and Utilit S l PV i l d Utility-Scale System Prices in the US: Current Drivers and Cost-Reduction Opportunities. National Renewable Energy Laboratory (NREL), February 2012. Alison King, and Margarett Jolly. Combining Permitting, Interconnection, and Incentive Applications: A New York City Case Study. NYC Solar City, 2011. www.nycsolarcity.com. Barbose, Galen, Darghouth, Naim, Wiser, Ryan, and Seel, Joachim. Tracking the Sun IV- A Historical Summary of Installed Cost of Photovoltaics in the United States from 1998 to 2010. Berkeley, CA: Lawrence Berkeley National Laboratory, September 2011. http://eetd.lbl.gov/ea/ems/reports/lbnl5047e.pdf. Board of Governors of the Federal Reserve System. Foreign Exchange Rates G.5, January 8, 2012. http://www.federalreserve.gov/releases/g5/. http://www federalreserve gov/releases/g5/ Brooks, Bill. Expedited Permit Process for PV System - A Standardized Process for the Review of Small-scale PV Systems. Solar America Board for Codes and Standards, October 2011. Bundesnetzagentur (BNetzA). Monthly PV System Interconnection Announcements 2009-2011, 2012. http://www.bundesnetzagentur.de/DE/Sachgebiete/ElektrizitaetGas/AnzeigenMitteilungen/MeldungPhot ovoltaikanlagen/MeldungPhotovoltaikanlagen_node.html. lt ik l /M ld Ph t lt ik l d ht l Bundesverband Solarwirtschaft e. V. (BSW-Solar). Statistische Zahlen Der Deutschen Solarstrombranche (PV). Berlin, 2011. http://www.solarwirtschaft.de/fileadmin/content_files/201105_BSW_Solar_Faktenblatt_PV.pdf. DSIRE. Solar Sales Tax Incentives. Database of State Incentives for Renewables and Efficiency, n.d. y, http://www.dsireusa.org/solar/solarpolicyguide/?id=12.

47

Bibliography g p y
EuPD. Database of Installer Offer Prices for German Residential PV Systems 2006-2011, 2012. Frank Stubenrauch. National Survey Report of PV Power Applications in Germany 2002. International Energy A E Agency (IEA) 2003. (IEA), 2003 German Statistical Federal Agency. Monthly Price Indices for Cost of Living and Index of Retail Prices 1991-2011, January 8, 2012. www.destatis.de. Gerrit Jan Schaeffer, Erik Alsema, Ad Seebregts, Luuk Beurskens, Hugo de Moor, Wilfried van Sark, Michael Durstewitz, et al. Learning from the Sun. Energy Research Centre of the Netherlands ( g gy (ECN), ) August 2004. GTM Research, and Solar Energy Industries Association (SEIA). US Solar Energy Trade Assessment 2011, November 2011. GTM Research. U.S. Solar Market Insight Report 2010. Solar Energy Industry Association (SEIA), 2011. . U.S. Solar Market Insight Report 2011. Solar Energy Industry Association (SEIA), 2012. Helmut Krmer-Eis. Perspektiven Erneuerbarer Energien Teil 1: Photovoltaik. KfW Beitrge Zur Mittelstands- Und Strukturpolitik 12. Frankfurt am Main: Kreditanstalt fr Wiederaufbau (KfW), November 1999. Hugh Bromley California Versus German Solar Prices: Same Dope, Twice as High Bloomberg New Bromley. Dope High. Energy Finance, February 2012. Haas, Reinhard. Progress in Markets for Grid-Connected PV Systems in the built Environment. Progress in Photovoltaics: Research and Applications. 2004(12), pp.427-440. IHS Emerging Energy Research. Europe Solar PV, Markets and Strategies 2011-2025, September 2011. 2011

48

Bibliography g p y
International Weather for Energy Calculations (IWEC). Insolation Rates for Hamburg, Germany. U.S. Department of Energy: EnergyPlus Energy Simulation Software, December 2011. http://apps1.eere.energy.gov/buildings/energyplus/cfm/weather_data3.cfm/region=6_europe_wmo_region_6/country=DEU/ cname=Germany. cname=Germany . Insolation Rates for Munich, Germany. U.S. Department of Energy: EnergyPlus Energy Simulation Software, December 2011. http://apps1.eere.energy.gov/buildings/energyplus/cfm/weather_data3.cfm/region=6_europe_wmo_region_6/country=DEU/ cname=Germany. IRENA. Renewable Energy - Power Sector Costing Study. International Renewable Energy Agency (IRENA), 2011. Joachim Seel. Survey of German Residential PV Installers, February 2012. Katie Bolcar, and Kristen Ardani. National Survey Report of PV Power Applications in the US 2010. Exchange and Dissemination of PV Power Systems. International Energy Agency (IEA), 2011. Klaus Oppermann. Perspektiven Erneuerbarer Energien Teil 4: Frdergebnisse Des 100.000 Dcher-Solarstromg Programms- Eine Zwischenbilanz. KfW Beitrge Zur Mittelstands- Und Strukturpolitik 28. Frankfurt am Main: Kreditanstalt fr Wiederaufbau (KfW), August 2002. . Perspektiven Erneuerbarer Energien: Das 100.000 Dcher-Solarstrom-Programm: Eine Schlussbilanz. KfW Beitrge Zur Mittelstands- Und Strukturpolitik 31. Frankfurt am Main: Kreditanstalt fr Wiederaufbau (KfW), November 2004. Kristen Ardani. Non-Hardware BoS Cost for PV Systems Database 2010. National Renewable Energy Laboratory (NREL), (NREL) February 2012. 2012 Dr. Lars Podlowski. Applying German Design Practices in the US. Solon, 2008. Lauren Poole, Paul D. Maycock, and Ward Bower. National Survey Report of PV Power Applications in the US 2007. Exchange and Dissemination of PV Power Systems. International Energy Agency (IEA), 2008. Langen, Christian. Complexity cost and economies of scale, why residential customers in Germany pay 25% for a PV system than US customers, SolarPower International 2010 Sovello AG customers 2010,

49

Bibliography g p y
Lothar Wissing. National Survey Report of PV Power Applications in Germany 2006. Exchange and Dissemination of PV Power Systems. International Energy Agency (IEA), May 2006. . N ti National Survey R lS Report of PV P t f Power A li ti Applications in Germany 2008 V i 2 E h i G Version 2. Exchange and d Dissemination of PV Power Systems. International Energy Agency (IEA), May 2009. . National Survey Report of PV Power Applications in Germany 2010. Exchange and Dissemination of PV Power Systems. International Energy Agency (IEA), May 2011. Marco Tepper. Statistical Data of the German Solar Electric Industry ( ) Bundesverband der pp y (PV). Solarwirtschaft (BSW), April 1, 2012. http://www.solarwirtschaft.de/fileadmin/media/pdf/bsw_solar_fakten_pv.pdf. Martin Junginger, Wilfried van Sark, and Andr Faaji. Technological Learning in the Energy Sector. Northampton, Massachusetts, USA: Edward Elgar Publishing, Inc., 2010. McFreely, David. McFreely David Study of Solar Permitting and Inspection A Study of Issues with Existing Procedures Inspectionand the Need for an Industry Standard. San Jose: SolarTech, September 2011. Melanie Persem, Thomas Chrometzka, Christian Brennig, Rainer Brohm, Frederik Moch, and Thomas Hielscher. Reduction of Administrative Barriers for PV Systems in Germany at the National Level. PV Legal. Bundesverband der Solarwirtschaft (BSW), 2011. Melanie Persem Thomas Chrometzka and Rainer Brohm Reduction of Administrative Barriers for PV Persem, Chrometzka, Brohm. Systems in Germany at the Regional Level. PV Legal. Bundesverband der Solarwirtschaft (BSW), 2011. Mills, Carl, and Newick, Kurt. Solar Electric Permit Fees in Northern California. Sierra Club, July 2011. Naim Darghouth, Galen Barbose, and Ryan Wiser. The Impact of Rate Design and Net Metering on the Bill Savings from Distributed PV for Residential Customers in California. Lawrence Berkeley National Laboratory (LBNL) A il 2010 L b t (LBNL), April 2010.

50

Bibliography g p y
National Renewable Energy Laboratory (NREL). National Survey Report of PV Power Applications in the US 2008. Exchange and Dissemination of PV Power Systems. International Energy Agency (IEA), 2009. . National Survey Report of PV Power Applications in the US 2009. Exchange and Dissemination of PV Power Systems. International Energy Agency (IEA), 2010. . System Advisor Model (SAM), n.d. https://sam.nrel.gov/. Paul D. Maycock, and Ward Bower. National Survey Report of PV Power Applications in the US 2002. y y p pp Exchange and Dissemination of PV Power Systems. International Energy Agency (IEA), 2003. Paul Gipe. Rate of Return Calculation of Solar PV Using Excel Rate Function, April 2007. http://www.wind-works.org/Solar/RateofReturnCalculationofSolarPVUsingIRATEFunction.html. Paula Mints. Capacity, Prices, and All That Solar Jazz. Renewable Energy World, March 2011. http://www.renewableenergyworld.com/rea/news/article/2011/03/capacity prices http://www renewableenergyworld com/rea/news/article/2011/03/capacity--prices-. Payne, Doug. Overcoming US Market Barriers to PV presented at the Intersolar, San Francisco, July 13, 2011. PHOTON Consulting, LLC. The Next Wave. Solar Annual, 2012. . The True Cost of Solar Power 2011 The Pressure Is On 2011 2011, On, 2011. Photovoltaic Solar Resource: United States and Germany. National Renewable Energy Laboratory (NREL), May 2008. http://www.seia.org/galleries/default-file/PVMap_USandGermany.pdf. Pitt, Damian. Taking the Red Tape Out of Green Power - How to Overcome Permitting Obstacles to Small-scale Distributed Renewable Energy. Network for New Energy Choices, 2008. PvXchange. pvX-Spotmarket Price Index for Modules, December 1, 2011. http://www.pvxchange.com.

51

Bibliography g p y
Reichmuth, Matthias. Vorbereitung Und Begleitung Des EEG Erfahrungsberichtes 2011- IIcSolare Strahlungsenergie. Berlin: Leipziger Institut fr Energie, June 2011. Renewable Analytics. Database of Survey Results 2010,2011, February 2012. Sprague, Ethan. The Impact of Local Permitting on the Cost of Solar Power. SunRun, January 2011. Statistisches B d St ti ti h Bundesamt. Fr 2011 Wi d Mit Ei t F Wird Einer L i ht B lk Leichten Bevlkerungszunahme G h Gerechnet, h t January 13, 2012. https://www.destatis.de/DE/PresseService/Presse/Pressemitteilungen/2012/01/PD12_014_12411. html;jsessionid=6DFB56EB33FF754CAC381C7CC029C18D.cae2. Susannah Pedigo, Paul D. Maycock, and Ward Bower. National Survey Report of PV Power Applications in the US 2006. Exchange and Dissemination of PV Power Systems. International Energy Agency (IEA), 2007. The Vote Solar Initiative. Streamlining the Solar Permitting Process, 2011. http://votesolar.org/cityinitiatives/project-permit/bestpractices.html. U.S. Census Bureau, Population Division. Annual Estimates of the Population for the United States, Regions, States, and Puerto Rico: April 1, 2010 to July 1, 2011 (NST-EST2011-01), December 2011. Wiser, Ryan, Barbose, Galen, Darghouth, Naim, and Seel, Joachim. Tracking the Sun Database: 1998-2011 1998-2011. Lawrence Berkeley National Laboratory (LBNL) 2012 (LBNL),

52

Appendix: Currency Conversion pp y

Inflationandexchangeratefactors
Germaninflation+2011xrate USinflationfactor 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 0.9 200 00 200 01 200 02 200 03 200 04 200 05 200 06 200 07 200 08 200 09 201 10 201 11 variablexrateandUSinflation Germaninflationfactor

FirstGermanpriceswerenormalizesfor2011,whichwerethenconvertedto$usingtheaverage exchangerateoftheyear2011of$1.39/.

53