learning curves for photovoltaics - gregory nemet

Learning Curves for Photovoltaics

Gregory [email protected]

University of Wisconsin – Madison

June 2007International Energy Agency

Gregory Nemet Learning Curves for PV 1

Learning curves for photovoltaics

Session 1 items:• latest rates• prices vs costs• structural changes• stability of l-rates• prediction• geographic scope• niche markets

2 points:1 If experience curves. . .

reliability of predictions2 If tech4 more generally. . .

expectations


Cost of electricity from PV


1. Reliability of experience curve projections

Experience curve for PV modules



Frequency distribution of learning rates calculated in 156learning curve studies.

Data: Dutton and Thomas (1984); McDonald and Schrattenholzer (2001);Nemet (2007)



Experience curve for PV modules



1985 1990 1995 2000 20050.1

0.15

0.2

0.25

Learning rates calculated for varying periods

Lear

ning

rat

es

End year of learning interval



0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.40

5

10

15

20

25

30

Calculated learning rates

Fre

quen

cy (

n=23

1)

Learning rate

5th to 95th percentile = 0.17 to 0.24



2010 2020 2030 2040 2050 2060 2070 20800

10

20

30

Year at which price of PV equals that of competing technology

Fre

quen

cy (

n=23

1)

Breakeven year

5th to 95th percentile = 2029 to 2040



0 100 200 300 400 500 600 7000

10

20

30

Cost to subsidize PV through breakeven

Fre

quen

cy (

n=23

1)

Billions of 2005 $s

5th to 95th percentile = $50b to $166b


2. Decomposition of cost-reductions

• Which factors were most important in reducing cost?• What were the primary drivers of change in those factors?


2. Decomposition of cost-reductions

Questions:• Which factors were most important in reducing cost?• What were the primary drivers of change in those factors?

Approach:1 Identify technical and economic factors2 Time-series data on each factor3 Model impact of factors4 Assess influence of experience, etc.


2. Change in each factor 1975-2001


2. Model results: contribution of each factor

Portion of cost reduction accounted for by each factor,1980-2001

$3.685%2%2%

3%12%30%

43%

3%

$25.30

$0

$5

$10

$15

$20

$25

200

2 $/

W

1979price

2001price

efficiencyplantsize

Siprice

wafersize

Siused

yield poly-x-stal

un-explained


2. Drivers of change in cost-reducing factors

Learning-by-doing (lbd) only weakly explains change in themost important factors. . . other drivers play a role

Factor Cost impact Drivers of change in each factor

Plant size 43% Expected future demandEfficiency 30% R&D, some lbd for lab-to-marketSilicon cost 12% Spillover benefit from IT industry

Wafer size 3% Strong lbdSi use 3% Lbd, but spillover for wire-sawsYield 2% Strong lbdPoly share 2% New process, lbd possibleOther factors 5% Not examined

Nemet, G. F. (2006). “Beyond the learning curve: factors influencing costreductions in photovoltaics.” Energy Policy 34(17): 3218-3232.


2. Did experience enable increase in plant size?

• Several firms make rapid expansions with little experience.• Biggest plants built by firms with access to capital.

0

5

10

15

1975 1980 1985 1990 1995 2000

Out

put p

er p

lant

(MW

/yea

r)

Managing risk of large, uncertain investments w/ lengthypayoffs more important than overcoming technical problems.


Summary

If we are going to rely on experience curves. . .1 Need to be explicit about reliability of predictions2 Policy decisions should reflect variation in L-rates

If we consider the drivers of technological change itself. . .• Need to create incentives for investments in

cost-reducing activities• Investments drive cost reductions• Expectations about future demand drive investment• Government activities affect these expectations• . . . and perceptions of risk


Appendix

Extra slides. . .


Experience curves to predict change in costs

(Wright 1936, Arrow 1962, Conley 1970)

Costt = Cost0

(qt

q0

)−b

q = cumulative capacity, b = learning coefficient

Advantages• Data availability• Model consistent with

narratives• High goodness-of-fit• Dynamic predictions• Single parameter

Limitations• Sensitive to timing• Discontinuities• Technical constraints• Highly aggregated• Other causal factors• 1-dimensional quality


Prices vs. costs: change in competition

1975-1979: from 2 firms to dozens.


learning curves for photovoltaics - gregory nemet

Documents