Hedging Strategy and Electricity Contract Engineering - IFOR

118 Power portfolio optimization
Production assets A power portfolio consists not only of financial and physical
contracts, hereby called contract portfolio, but also of production assets,
hereby called production portfolio. Numerous of the power plants
exhibit operational flexibility, as discussed in Chapter 4, where we concluded
that each plant type corresponds to a specific set of contracts. By
viewing production facilities as contracts, it is natural to state that an optimal
power portfolio implies not only an optimal contract portfolio, but
also an optimal production portfolio. The production portfolio can be optimized
on two levels. The first level is to find the optimal use of the
operation flexibility, i. e. to find the optimal dispatch strategy for the existing
plants. The second level is to find the the best portfolio of power
plants, i. e. to allow for acquisitions and sell-offs of these assets.
Complex contracts The contracts in the power market can be very complex
in comparison with traditional financial contracts. A major difference is
that many OTC power contracts have an uncertain underlying quantity
of electricity as described in Chapter 2.7. In addition to the price risk
that players in the traditional financial markets are facing, the electricity
players often also face a volume risk stemming from swing options or
from outages of plants. Such volume uncertainty cannot be handled by
the simple Markowitz approach (3.1), but naturally needs to be dealt with
by a power portfolio optimization.
Simultaneous optimization Holthausen [Hol79] stated a separation theorem
that production scheduling can be done independently from hedging, under
the assumption of no production uncertainty and no basis risk. In
the electricity market there is generally a presence of basis risk, due to
the finite number of nodes in the grid at which derivatives are traded,
and definitely a production uncertainty, why the separation theorem does
not hold. Anderson & Danthine [AD80] showed that production must
be determined jointly with hedging in the case of basis risk. Viaene &
Zilcha [VZ98] have showed that even in the absence of basis risk the separation
theorem breaks down in the case of correlation between input costs
and output price. This correlation exists in the electricity market through,
for example, the correlation between fuel prices or water inflow and electricity
prices. Thus we can conclude that a simultaneous optimization of
the production and the contract portfolio, which could typically be used to
hedge future profits from the production portfolio, is needed. It is obvious