DRAFT
MEMORANDUM
To:
Paul
Augustine,
EPA/
CAMD
From:
Barry
Galef
and
Bansari
Saha,
ICF
Consulting
Subject:
Alternative
SO2
Allowance
Allocation
Approaches
under
CAIR
Date:
March
9,
2005
This
memorandum
responds
to
EPA's
request
for
a
short
write­
up
comparing
SO2
allowance
allocation
outcomes
under
the
existing
methodology
in
CAIR
with
other
allocation
mechanisms.
The
existing
allocation
mechanism
is
based
on
Title
IV
allocations,
using
fuel
input
data
from
the
1980s.
Commenters
to
the
CAIR
rulemaking
have
raised
questions
regarding
the
reasonableness
of
using
such
an
old
baseline
as
opposed
to
more
recent
baseline
heat
input
data.
This
memorandum
argues
that
the
final
market
outcomes
based
on
marginal
cost
considerations
in
IPM
will
not
be
affected
by
using
more
recent
heat
input
data
to
allocate
the
allowances.

The
effects
of
using
more
recent
baseline
heat
input
data
can
be
approached
from
two
related
perspectives.
First,
what
are
the
theoretical
implications
of
changing
the
existing
SO2
allowance
allocation
mechanism
under
CAIR
to
a
different
input­
based
allocation
mechanism
in
which
EPA
uses
more
recent
heat
input
data?
Second,
how
would
EPA's
analysis
using
IPM
change
in
practice,
in
response
to
different
allocation
mechanisms?

The
answer
to
the
first
question
depends
on
whether
we
are
interested
in
the
distribution
of
profits
and
losses
resulting
from
CAIR,
or
only
in
the
physical
effects
of
the
program.
The
choice
of
the
baseline
year
will
certainly
affect
the
initial
allocation
of
allowances,
in
terms
of
the
number
of
allowances
allocated
to
different
plants.
Although
the
initial
allocation
mechanism
does
not
help
or
hurt
states
and
industries
in
the
aggregate,
any
allocation
creates
winners
and
losers,
and
changing
the
initial
allocation
mechanism
will
change
the
identities
of
those
winners
and
losers.
Allowances
have
significant
value
and
the
initial
allocation
determines
who
gets
these
instruments
of
value.

In
theory,
though,
changing
the
baseline
will
have
virtually
no
real­
world
effects
other
than
the
distribution
of
winners
and
losers,
so
long
as
the
regulatory
program
uses
a
permanent
allowance
allocation
system
and
efficient
market
mechanisms.
1
This
conclusion
follows
from
the
fact
that
allowances
have
intrinsic
value,
and
their
holders
know
that
an
allowance's
value
is
determined
by
its
value
to
potential
buyers,
not
by
its
original
holder.
The
value
to
potential
buyers,
in
turn,
is
determined
by
the
allowance's
ability
to
save
its
purchaser
the
cost
of
reducing
emissions
by
a
given
amount.
Thus,
assuming
minimal
transaction
costs
in
trading
allowances
and
complete
information
flow
in
the
allowance
market,
allowances
will
be
traded
until
marginal
abatement
costs
are
virtually
equalized,
and
the
same
final
distribution
of
1
A
permanent
allocation
system
is
one
where
allowances
are
allocated
based
on
a
fixed
or
constant
formula
and
the
initial
allocation
never
changes
(
or
changes
in
ways
that
are
specified
in
advance
 
such
as
a
fixed
reduction
every
year,
so
as
not
to
induce
any
strategic
behavior
from
affected
entities).
DRAFT
CAIR
Allowance
Allocation
allowances
will
result
regardless
of
the
initial
allocation.
If
the
market
reaches
the
same
final
distribution
of
allowances,
it
also
reaches
the
same
control
decisions,
emissions,
prices,
and
total
costs,
independent
of
the
allocation
mechanism.

In
practice,
EPA
analyzes
the
costs
and
other
effects
of
cap­
and­
trade
programs
like
CAIR
using
IPM.
IPM's
analytical
approach
is
to
find
the
most
efficient
way
to
meet
a
given
emission
cap,
and
implicitly
assumes
that
the
market
(
acting
through
a
well­
functioning
allowance
trading
system)
will
ensure
that
the
cap
is
indeed
met
efficiently.
Because
IPM
implicitly
assumes
that
the
allowance
trading
mechanism
works
well
enough
to
direct
the
allowances
to
the
right
users,
it
does
not
use
any
information
on
allowances,
and
is
therefore
insensitive
to
the
initial
allowance
allocation.
Essentially,
IPM
treats
the
allocation
as
a
sunk
cost
(
or
sunk
benefit)
that
does
not
affect
decisions
at
the
margin.
Hence,
changing
the
baseline
for
allocating
allowances
will
not
affect
IPM
results
on
compliance
decisions,
wholesale
electricity
prices,
or
allowance
prices,
or
net
compliance
costs.

Thus,
as
long
as
the
allowances
are
allocated
based
on
a
permanent
allocation
mechanism,
changing
the
baseline
year
for
the
heat
input
data
used
to
calculate
the
initial
allocations
will
not
significantly
alter
the
market
outcomes
such
as
electricity
prices,
compliance
decisions,
or
retirements,
under
the
assumption
of
an
efficient
trading
system
with
a
free
flow
of
market
information.
Though
differences
in
the
allocation
baseline
could
affect
the
distribution
of
gains
and
losses,
it
will
not
affect
IPM
results
because
the
model
determines
the
solutions
related
to
marginal
cost
calculations
under
an
efficient
trading
system.
