Technology Choices in the U.S. Electricity
Industry before and after Market
Restructuring
IAEE Singapore
Zsuzsanna Csereklyei & David Stern
Introduction
• We study the drivers of electricity generation technology adoption
between 1970 and 2014 in the lower 48 U.S. states.
• Since the 1990s, major electricity market restructuring took place in
some parts of the United States.
• We explore the implications of changing from a regulated “cost-ofservice” or rate of return system to a partly or fully deregulated
market on technology and fuel choices.
2
Our questions:
• What drives investment choice of certain type of power plants,
defined as a combination of
1) fuel and
2) technology (efficiency, peak vs. baseload)?
• In particular, how do own and substitute fuel prices impact on the
choice of fuel and the efficiency level of the new plant (under
regulated vs. wholesale markets)?
• How does access to wholesale markets influence this choice?
3
0
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
MW Capacity Addi ons
Power Plant Investments 1970-2014
70000
60000
50000
OWT
SPV
40000
NUC
CHYD
CIGCC
30000
CSC PC
CSC FB
NGST
20000
NGFCT
NGFCC
10000
4
Background
•
•
•
•
•
US electricity market was characterised by a cost-of-service model before
restructuring  regulated return on investment over the prudently incurred
operating costs.
US Energy Policies: PURPA 1978 (public utilities introducing IPP).
Natural Gas Policy Act 1978  Natural Gas Wellhead Decontrol Act 1989 
drop in natural gas prices. 1992 FERC O 636required gas pipeline companies
to open to all transporters fall in natural gas prices
Wholesale electricity market liberalization: driven by the difference in marginal
vs. average costs late 1990s, 2000s, in states where economic incentives existed.
FERC orders 888/889/2000
5
ISO/RTO Organization / North America
•
Six of the seven
ISO/RTOs started as
preexisting power pools
supported by states but
regulated by federal
laws.
•
In wholesale markets
natural gas prices, even
though only a part the
production portfolio
tended to determine the
marginal cost of
electricity supply.
Source: https://www.ferc.gov/industries/electric/indus-act/rto.asp
6
0.00
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Real 2009 dollars per million Btu
Real coal and natural gas prices delivered to the power sector
10.00
9.00
8.00
7.00
6.00
5.00
coal
4.00
natural gas
3.00
2.00
1.00
7
Assumptions
Factors determining investment decision:
• Long run-demand, measured by GDP development
• Capacity additions and retirements before construction (or during)
• Average fuel prices before construction: own and substitute fuel prices  over
80% of operating cost for coal/gas plants. Marginal costs = (fuel cost) is
impacting on dispatch ranking and determines the electricity price margin
• Efficient Market Hypothesis (current prices vs expected prices)
• Construction time duration (EIA and OECD data)
8
Model 1
Our basic model for each technology type:
9
Data
•
Generator Data: EIA 860 Form: All generators from 1970 to 2014 that were
connected to the grid and operational or retired.
•
•
•
•
The EIA distinguishes over 20 technologies, and several subtechnologies for
conventional steam coal.
US GDP (1970 -2014) from the BEA.
Coal, natural gas prices in the electricity generating sector from the EIA State
Energy Data System (SEDS).
Partly (wholesale only) or fully (wholesale and retail) restructured market
dummies after Craig and Savage (2013).
10
Model 2
11
Robustness checks
• We also worked with different construction times (EIA construction
times)
• We also added different (shorter) lags of the additions/retirements to
account for plants in construction in a state during the investment
decision.
12
Results
13
Results
• Positive effect of the liberalization of wholesale markets on the
construction of natural gas combined cycle technologies, and a negative
effect on a wide range of steam coal technologies, however high
efficiency coal generators (including fluidized bed plants) are less
negatively influenced by market liberalization.
•
The incremental impact of fully liberalized markets is generally not
statistically significant.
14
Results – Natural gas own fuel price elasticity
• The own price elasticity of natural gas plants is negative and significant
for combined cycle generators in non-liberalized markets.
• In wholesale markets the elasticities are highly significant and more
negative for both gas technologies.
• The coefficient for combined cycles is -1.67 and for gas turbines is 1.92, indicating a very elastic response in investment to gas price
changes. Both coefficients are significant at the 1% level.
15
Results – Coal fired power own fuel price elasticity
• The own fuel price elasticity of coal is negative but not statistically
significant for pulverized coal technologies in non-liberalized markets.
In liberalized markets the effect is positive but only significant for
subcritical plants at the 10% level.
16
Results – Cross price elasticities
•
The cross price elasticities of natural gas and coal technologies are not
statistically significant in non-liberalized markets.
•
The cross price elasticities of coal technologies with respect to the price of
natural gas in liberalized markets however show a clear investment substitution
towards coal fired baseload generation in case of rising natural gas prices.
•
The combined coefficients vary between 0.91 for supercritical and 1.02 for
subcritical technologies and are significant at the 5% level. This would indicate
about 1% increase in new coal capacity investment as a response to 1% increase
in natural gas prices.
17
All natural gas and coal fired new capacities in the U.S. as a % of the
total 1970-2014.
18
Results
• Our results indicate that while market liberalization per se has increased
investment flows into natural gas-fired plants, higher natural gas prices
under competitive regimes acted to reduce these investments, as the
relative profitability of other generation increased due to higher baseload
usage, and higher margins.
• Also, while the average impact of liberalization was negative on coalfired generators, natural gas prices under wholesale conditions promoted
shifting investment from natural gas to other forms of electricity
generation including renewables.
19
Conclusion / Implications
• Wholesale markets have significant impact on the choice of fuel and
some impact on the choice of technology.
• Natural gas supply and prices have the potential to significantly shape
the power generation landscape of states with wholesale electricity
markets in future.
• However, strongly falling renewable generation costs have to be taken
into account in the future.
• Price development on wholesale electricity markets
20
Thank you!
21