►
Description
In this Governance and Risk meeting, Steven Becker (Head of Risk), was joined by Rune Christensen (CEO), Lev Livnev (Economic Research Lead) and Greg Diprisco (Head of Business Development), as we dedicated the call to a review of the recent Stability Fee change.
This weeks agenda included:
- A review of what the Stability Fee is
- How it works
- How it will be implemented
- A brief discussion about the 'Impossible Trinity'
- Question and Answer session with the community
Stability Fee FAQ: https://www.reddit.com/r/MakerDAO/comments/93adqj/faq_stability_fee_raise/
Governance and Risk Framework - Part 1: https://medium.com/makerdao/makerdao-governance-risk-framework-38625f514101
Foundation Proposal: https://medium.com/makerdao/foundatio...
Website: https://makerdao.com
Twitter: https://twitter.com/makerdao
Reddit: https://www.reddit.com/r/MakerDAO/
Chat: https://chat.makerdao.com/home
Email: info@makerdao.com
A
A
We're
going
to
be
discussing
it
in
detail,
we'll
be
talking
about
a
high-level
overview
about
what
it
is,
we're
going
to
dig
into
some
of
the
reasoning
behind
the
proposal
and
eventually
get
to
some
implementation
and
go
in
its
mechanics
issues.
Happily,
we
have
some
heavyweights
in
the
meeting
today,
so
we'll
go
around
the
room
and
introduce
or
get
them
to
introduce
them
so
Steven.
You
want
to
kick
us
off
know
a
bit
about
yourself.
A
A
B
A
If
you
want
to
kind
of
follow
along
with
it,
do
individual
sections-
and
maybe
some
research
last
week
that
that
could
be
useful
but
to
get
things
started.
I
think
that
Steven
it'd
be
great.
If
you
can
maybe
give
us
a
high-level
overview
of
what
the
stability
fee
is
all
about,
and
maybe
they
go
into
some
details.
B
So,
for
intents
and
purposes,
a
stability
fee
can
be
seen
as
an
interest
rate
and
what
we
need
to
do
is
figure
on
and
so
an
interest
rate
there
actually
aspects
to
it.
The
first
one
is
an
inflation
component,
I
mean
if
you're
going
to
learn
a
certain
amount
of
money
outs.
The
return
that
you
need
to
get
from
it
needs
to
cover
the
purchasing
power
that
you
lost
through
the
process
of
inflation,
so
there's
an
inflation
component.
The
second
component
is
obviously
with
respect
to
who
you
lending
to,
and
that
is
the
credit
component.
B
This
is
the
the
component
that
covers
the
cost
to
providing
a
service.
This
is
what
banks
will
have
calculated
and
included
into
their
general
interest
cost
to
their
retail
customers,
and
within
that
you
also
have
a
policy
tool
which
allows
you
to
adjust
the
overall
mechanism
to
allow
the
integrity
of
your
system
to
operate.
So
let
me
just
go
back
to
the
credit
component.
B
If
the
collateral
for
any
particular
situation
or
any
/,
you'll
appropriately
MPR
token
holders
have
to
step
up
and
cover
the
loan,
so
that
premium
is
for
that
particular
purpose.
Now
the
liquidity
preference
component
does
not
apply
to
us
because
all
loans
are
open-ended,
so
we
can
just
describe
that
altogether.
So
the
three
components
that
we
need
to
focus
on
is
inflation
credits
and
the
operational
components
and
for
today's
meeting
I
think
it's
the
operational
components
with
respect
to
how
it
is
used
as
a
policy
to
which
is
going
to
be
in
a
very
important.
A
B
B
B
But
first,
let's
just
have
a
look
at
you
know.
You
know
effectively
why
we
would
need
one
I
mean
sure.
The
the
unlined
aspect
here
is
wide
have
a
policy
tool
to
begin
with,
and
you
always
need
a
policy
tool
because
there
are
always
gonna
be
frictions
in
the
markets.
Now,
what
do
I
mean
by
frictions?
It
can
range
from
you
know,
just
generally
the
cost
of
transactions
all
the
way
through
to
implication
of
tax.
B
They
are
always
going
to
be
these
prohibitive
components
that
are
not
going
to
allow
everything
to
work
continuously
and
perfectly,
as
designed
by
you
know,
folks,
in
their
mind
and
in
their
models.
So
a
policy
tool
should
support
the
governing
dynamics
of
a
system
and
what
I
mean
by
that
is
that,
if
you
would
imagine
for
a
second
that
your
governing
dynamics
in
your
economy
are
working
perfectly,
then
what
you
presume
in
is
everything
is
continuous
in
this
friction
and
everything
flows
appropriately.
B
But
because
there
are
these
frictions,
your
governing
dynamics
will
need
to
be
supported
by
another
tool
in
the
event
that
those
frictions
arise
and
the
economy
is
not
as
robust
or
broad
and
as
deep
as
you
hope
it
would
be.
So
as
a
combination,
you
would
use
a
tool
to
react
to
the
states
of
the
economy.
B
If
the
economy
is
mature
and
robust,
then
you
probably
won't
need
to
use
the
tool
that
often,
but
if
it
is
new
and
it
still
hasn't
developed
appropriately,
then
it
would
be
important
to
consider
a
tool
along
with
your
your
governing
dynamics.
So
in
this
instance,
if
you
have
a
look
at
you
know
the
federal
reserve,
it's
a
very
simple
concept:
they
use
an
interest
rate
as
one
of
the
policy
tools.
B
They
also
have
other
open
market
operations
where
they
buy
and
sell
assets
in
large
scale,
but
principally
the
idea
is
that
that
interest
rates
rule
is
used
and
is
the
Open
Market
Committee
focuses
on
that
tool
and
thinks
about
it
on
a
monthly
basis
and
then
reports
pretty
much
on
a
monthly
basis,
know
what
they're
going
to
do
and,
generally
speaking,
that
tool
isn't
used
as
often
as
everyone
thinks
it
should
be
in
our
city.
Looking
at
an
economy
where
you
know
it's
developing,
the
the
blockchain
economy
is
brand-new.
It's
developing.
B
D
D
The
magnet
prize
have
died
towards
the
target
place
and
then
very
basic
logic
of
net,
and
so
this
is
like
a
comment
like
it's
like
a
policy
decision
that
it
that
sort
of
guess,
based
on
that,
like
fundamental
logic
of
if
you
increase
the
stability
fee,
you
can
use
that
to
push
the
price
up,
and
this
is
a
part.
I
was
not
talking
about
I've,
been
riding
all
over
ready
these
couple
of
days
right,
it's
like
it's,
some
pretty
well
I
mean
like
everything
in
in
like
finance
and
monetary
economics,
whatever
it's
like.
B
D
That
it's
supply
and
demand
I'm
like
if
you
increase
the
cost
to
borrow
less
people,
are
gonna
borrow
because
they're
gonna
move
elsewhere
to
borrow
from
other
places
or
borrowing
from
CDP's,
for
instance,
and
that
reduces
the
like.
Theoretically,
that
reduces
the
supply
right,
at
least,
if
I
can
sum
some
negative
effect
on
supply
over
time,
because
a
percentage
of
the
people
would
be
borrowing
for
maker
would
be
book
would
be
borrow
something
else
somewhere
else
instead
and
the.
D
Like
and
then
that
reduce
supply,
and
then
we
got
supply
and
demand
micro
economics
of
how,
if
you
reduce
supply,
you
can
and
you
keep
the
man
steady,
you
can
increase
the
price
or
in
our
case,
it's
more
about
like
it's,
about
reducing
the
growth
of
supply
relative
to
the
corrosive
demand
because
what's
been
going
on-
and
this
is
kind
of
what
we
already
been
we
already
published
is.
First
of
all,
this
bin
is
like
small
instability
in
the
market
price
of
dye
that
people
can
actually
look
if
they
go
on
home.
D
So
so
so
it's
actually
like
there
is
this,
like
this
general,
a
buildup
of
dye
inventory
by
people
who
don't
actually
want
to
hold
that
they
just
want
to
buy
and
sell
it
and
eventually,
if
they,
their
inventory,
keeps
building.
Eventually,
that
will
have
like
a
very
like
direct
effect
on
the
prices
in
the
market,
because
they
would
potentially
unloading
the
toriel
just
stop
buying
it
entirely,
and
so
what
we.
So?
What
everybody's
saying
is
we
want
to
reverse
this
trend?
D
A
Sorry
I
wanted
to
jump
in
here
and
just
ask
a
question
so
we're
talking
about
arbitrage,
er
is
sort
of
smoothing
up
the
market,
creating
some
more
efficiency
for
us
and
how
that
hasn't
been
happening
due
to
a
glut,
I
guess
with
CVP
creation
with
cheap
debt.
What
I'm
kind
of
curious
about
is
why
haven't
we
seen
a
behavior
where
CDP
owners
are
taking
advantage
of
cheap
dye
in
order
to
pay
down
debts?
D
A
D
So
I
think
we
had
like
we
had
like
what
people
don't
realize
this.
We
have
seen
this
and
it
happens
all
the
time
and
there
are
people
out
there
doing
that
right
now,
it's
just
like,
because
people
do
it
like,
because
markets
are
not
perfect.
Right,
like
there
are
people
out
there
following
these
incentives,
but
they
are
not
like
massively
capitalized.
You
know
like
it's
the
people
doing.
A
D
I
mean
diamonds,
so
I
guess
our
it's
like
first
well
know.
The
main
thing
is
really
that
too
many
people
like,
like
we've
been
the
the
CEP
system,
has
been
lending
out
money
to
cheaply
for
too
long
and
as
a
result,
there's
been
an
imbalance
of
people,
borrowing
diamond
system,
and
then
he
will
buy
died
from
those
who
are
sort
of
the
modern
makers
in
between
the
borrowers
and
the
holders.
And
we
want
to
slowed,
we
don't
want.
D
D
This
is
like
this
was
communicated.
It
wasn't
communicated
very
well,
it
was
like
kind
of
it
was
too
late.
He
kind
of
came
out
and
said
that
we
have
this
like
this
buildup
of
this
imbalance
in
the
economy,
but
we
are
completely
doing
like
I
mean
what
we're
doing
is
just
like
we're
following
what
the
plan
is
to
do
in
this
situation,
because
you
can
never
have
it's
never
possible
for
someone
to
set
and
interest
rate
or
stability
to
be
at
some
fixed
value
and
to
be
like.
D
That's
it
that's
what
people,
because
that
the
reason
why
we
have
seen
too
much
growth
in
the
supply
thought
relative
to
the
growth
in
supply
and
demand
worth.
I
is
that
people
are
moving
the
loans
from
elsewhere
over
to
maker,
because
we
achieve
our
than
those
other
places.
Right
so
like
is
the
load
rates
on
like
helping
people
from
when
they
make
a
decision.
They
go
with
CDP's
and
it's
just
been
too
good
a
deal.
D
Basically
so,
like
too
many
people
have
been
making
that
decision
and
at
the
same
time,
if
you
look
at
the
the
incentives
for
holding
that
there's
less
people
who,
like
move
from
the
bank,
account
over
to
buy,
maybe
because
in
the
mayhem
they
can
actually
get
an
interest
rate,
for
instance,
right
like
what
they
can.
There's
some
other
like
it's
more.
You
know
like
the
like
die
itself
is
still
not
like.
D
It
has
a
lot
of
cheek
like
it
has
a
significant
matching
in
a
problem
that,
of
course,
we
inferred
is
now
working
very
hard
to
solve,
and
we
have
these
like
extremely
cool
things
coming
out,
such
as
trade
shift
Church
of
cash,
and
these
dudes
have
a
financial
product
that
fundamentally
it
runs,
and
by
and
means
that
you
have
to
have
done
in
order
to
use.
But
before
we
get
to
that
point,
we
just
have
to
like
stabilize
what's
going
on
right
now
and
yeah.
D
Reaction
comes
from,
like
the
lack
of
expectations
said
on
suddenly
it's
this
coming
right,
like
we
didn't,
discuss
it
beforehand,
and
then
also
because
of
that
it
feels
like
it's
a
big
increase
when
really
what
from
I
want
to
be
smaller,
like
the
original
stability
fee
was
just
very
like
it's
not
that
we
want
to
yeah.
You
know
right.
We
just
consider
the
original
stability
for
you
to
be
very.
F
A
E
Criminal,
so
I
can
go
ahead
and
help
articulate
some
of
the
questions
that
have
been
asked
in
the
chat
and
I
can
actually
go
and
do
that
right
now.
So
rune
you've
basically
been
summarizing
an
answer
to
this,
but
Jordan
Crandall
asked
about
eight
minutes
ago.
How
significant
is
the
risk
of
gloss
liquidity
from
increasing
the
stability
fee
compared
with
the
risk
of
slipping
from
the
peg?
D
D
Mean
planning
so
the
low
liquidity
right
now
does
not
in
any
way
come
from
like
it's.
The
quiddity
is
not
constrained
on
the
supply
side.
It's
constrained
in
the
debate
so
like
it
doesn't
matter
because
all
the
baigan
vegas
ton
about
anyway.
So
they
don't
care
about
being
able
to
see
peas
right
now,
because
they
they
already
have
all
the
dye
they
need.
So
like
increasing
the
stability
people.
That
means
nothing.
What
it
doesn't
mean.
E
D
They
actually
like
they
don't
have
to
buy
as
much
diet
anymore
going
forward,
because
there
will
be
most
people
generally
buy
potentially,
and
maybe
they
can
also
sell
some
of
that
dye
to
people
who
go
to
the
closest
gdb
positions
and
in
general
it.
But
what
we're
hoping
is
this?
Will
this
will
change
expectations
in
the
marketplace
and
maybe
increase
faith
in
the
one
in
the
one
UST
target
price
and
as
a
result,
that
should
be
something
that
increases
the
query,
because.
D
Of
uncertainty
around
where
the
talk
like
where
the
dye
price
is
going
in
the
future
and
whether
there's
a
chance,
you
can
like
slave
away
from
the
pig
for
a
while.
You
could
like
pets
like
software.
Some
weaknesses
like
having
any
kind
of
uncertainty
like
that
significantly
reduces
how
comfortable
market
makers
are
and
creating
liquidity.
E
C
Mean
have
you
got
answer
that
I
think.
Can
you
guys
hear
me?
Okay
by
the
way,
yeah
okay,
great
I
would
say
that
a
I
think
died
USD
pairs
in
it,
I
USD
pairs
that
users
can
access.
So
you
know
because
USD
pairs
require
Fiat
banking
infrastructure.
There's
no
such
thing
as
a
u.s.
die
USD
pair
that
everyone
can
access,
because
it'll
be
out
depending
on
the
terms
of
a
centralized
exchange
and
geography
that's
available
and
who
is
eligible
to
open
such
accounts
and
use
them
and
so
on.
C
C
However,
it
doesn't
ultimately
address
like
systemic
supply
imbalance,
so
supply-demand
imbalance
problems.
It's
it's
more
of
it.
It
resolves
kind
of
sort
of
you
can
think
of
them
as
market
structure
and
liquidity
issues
or
technical
liquidity
issues
that
arise
from
the
difficulty
of
making
markets
in
on
volatile
volatile
pairs
more
so
than
addressing
a
systemic.
Supplying
the
men
in
balance-
because
you
know
I
mean
the
eath
market-
is
fairly
liquid.
C
So,
ultimately,
if
market
makers
were
ending
up
with
with
with
too
much
ether
or
something
I
thought
think
they
could
always
shift
that
into
any
other
asset,
including
the
USD.
So
in
the
in
the
grand
scheme
of
things
that
doesn't
actually
resolve
a
problem
like
that,
but
it
does
make.
The
kind
of
thing
does
make
the
trading
much
more
technically
efficient,
which
is
great
for
the
quick
to
let.
A
C
A
That
it's
it's
sort
of
this
cigarette.
We
can
have
this
sort
of
like
this
time.
I
go
happening
here,
but
would
have
different
competing
groups,
but
I'm
wondering
if
there's
a
chicken-and-egg
situation
going
on
where
we
need
to
stabilize
supply
and
demand
and
strengthen
the
peg
in
it
in
advance
of
multi
collateral
dying
when
multi
collateral
die
arrives.
Some
of
the
challenges
that
we're
facing
with
you
know,
velocity
and
liquidity
will
probably
be
solved
at
that
point.
Was
that
very
distant.
A
Yeah
I
think
I'm
trying
to
figure
out
so
we're
strengthening
the
peg,
we're
creating
some
some
additional
incentives
to
even
up
the
supply
and
demand
pressures
that
we
have,
but
it's
the
the
demand
side
that
is
lacking
at
this
moment
and
presumably,
if
multiplied
will
die
when
the
gloves
come
off
and
we
we
have
much
higher
debt
ceilings,
we
have
greater
adoption.
We
have
large
partners
like
trade
ship
using
die.
You
know
in
their
day
to
day
business.
Presumably
the
supply
or
the
demand
side
will
be
taken
care
of
at
that
point.
Well,.
C
Yeah,
it
would
look
completely
different
as
what
we
can
say
as
well.
Of
course,
you
know,
as
the
system
scales
will
have
many
more
users
on
all
sides
of
this,
of
this
thing
and
it's
hard
to
say
how
exactly
they
will
they
will.
They
will
balance
out
and
I
think
it's
very
hard
to
make
a
prediction
that
you
know
at
a
certain
date.
There
will
be
enough.
There
will
be
so
many
users
on
this
side
and
someone
users
on
this
item,
it'll
be
balanced.
C
It's
unfortunately,
we'll
kind
of
have
to
see,
and
things
will
will
develop,
I.
Think
unexpectedly,
so
you
don't
always
know
when
you'll
get
suddenly
a
lot
more
CVP
users.
Some
people
get
a
lot
of,
basically
speaking,
it's
actually
mostly
I
would
say
completely
out
of
control
of
the
maker
community.
So
if
tomorrow
some
some
some
DAP
came
out,
that
was
really
popular
and
it
was
using
dye
as
its
main
unit
of
account
and
people
were
now
buying
dye
in
order
to
be
able
to
trade
on
it.
C
That
would
be
great
news,
because
this
is
the
sort
of
thing
that
we
are
here
for,
but
that
that
could
appear
and
then
suddenly
you
will
have
a
huge
demand
for
dye
coming
from
people
who
want
to
use
this
application
and
nothing
to
do
with
any
of
our
any
of
our
policies.
So
the
thing
that
I'm
trying
to
stress
is
actually
that
policy
needs
to
be
adjusted
to
suit
circumstance,
rather
than
the
other
way
around.
So
it's
there
will
always
be
the
they're
all
going
to
be
these
realities
of
of
how
dye
is
being
used.
C
That
will
have
some
influence
on
how
risk
parameters
have
to
be
set
and
I
think
this
is
well.
This
is,
if
we
kind
of
bring
up
this
idea
of
the
impossible
Trinity.
This
is
actually
the
sort
of
thing
that
it's
saying
is
that
you
can't
you
can't
independently
control
all
three
of
those
variables
that
if
there
are
some
some
kind
of
exogenous
force
affecting
the
supply
and
demand
for
dyes,
so,
for
example,
just
you
know
some
market
sentiment
which
affects
how
many
people
want
open,
CDP's
or
who
else
is
you
know
lending?
C
You
know
in
a
similar
in
a
similar
fashion
to
CDP.
So
actually,
if
you
know
there's
just
some
other,
so
they
say,
there's
a
centralized
exchange,
that's
lending,
that's
providing
margin
loans
to
its
users.
That
will
have
some
effect
on
on
the
diary
ecosystem
because
it
will
it's
AB,
it's
another
competitive
player
in
the
market
and
then
users,
you
know
in
an
ideal
market-
should
choose
the
one
that
offers
the
most
attractive
terms.
C
C
We,
it
wasn't
driven
by
us,
seeing
that
we
want
to
charge
more
for
loans
collateralized
by
Keith,
because
we're
seeing
more
risk,
and
even
we
were
seeing
before
it
wasn't
coming
from
from
from
that
consideration.
It's
coming
from
this
from
the
second
type
of
consideration
which
I
think
which
is
or
which
is
a
real
one.
C
E
D
Is
this
because
I
basically
talked
about
this
before,
and
so
it's
not
the
reason
why
it
appears
like
it's.
Such
a
dramatic
change
is
more
because
the
initial
value
was
very
dramatically
low
and
the
reason
why
we're
so
low
was
because
we
expect
like
because
the
initial
user
experience
of
CDP's
basically
was
so
incredibly
complicated
that
you
fell.
There
was
a
secret
like
we
expected.
D
There
would
be
so
many
such
a
significant
problem
with
using
it
that
we
just
had
to
set
this
ability
for
super
low
to
compensate,
and
now
that
I
mean
now
the
people
really
getting
more
used
to
it
and
don't
worry
I
mean
it's
actually
getting
a
massive
upgrade
and
we're
having
much
easier
and
stuff
in
the
future.
But
but.
D
The
parent
interface
right-
it's
just
like,
like
people,
people,
definitely
use
like
anyone
who
knows
about
it
like
they're,
not
really
being
held
back
by
the
interface
to
you
know
for
them
to
get
their
super
cheap
loans.
So
basically
it's
just
then
like
we
have
right
like
what
we
are
observing
is
as
a
result
of
offering
super
super
cheap
lows
that
are
just
like
outrageously,
cheap
because
of
this,
like
extra,
like
this
original
factor
of
the
user
experience
being
very
harsh,
and-
and
so
so
it's
like
it's
like.
D
We
know
for
sure
that
I
mean
we
know
it's
caused
by
super
super
cheap
loans
so
like.
If
we
just
like,
we
can
increase
the
loan,
but
it's
like
whatever
keep
it
in
Super,
G
below
and
land.
It
doesn't
really
matter
because
it'll
still
just
be
like
a
super,
ridiculously
cheap
below
the
people
will
continually
use
any
time
they
can
like
for
this.
You
know
like
so
it's
kinda
I
mean
it's
it's
that
we.
If
we
really
expect
you
wouldn't
possibly
be
able
to
even
measure
the
impact
of
ello
increase.
D
D
Were
just
we
just
fully,
don't
believe
that
it
would,
it
would
rectify
the
imbalance,
which
is,
of
course
very
hard
to
determine,
but
it's
just
like
I
mean
remember.
We
don't
just
want
a
supply
of
dye
like
growth
in
the
supply
by
to
line
up
with
growth
in
demand.
You
actually
want
growth
of
the
supply
and
by
to
go
slightly
below
the
demand
curve,
like
the
growth
and
demand
thought
so
so
like
that
basically
speaks
to
do
it.
D
D
You
know
we
were
like.
It
was
like
a
trick
right
where
we
like
convince
them
that
it's
not
gonna,
be
a
big
change
and
then
you
still
end
up
with
the
same
yes,
I
mean
this
we're
gonna
in
there
anyway
know
what
this
whole
thing
is
all
about
the
impossible
to
me
right
and
let's
just
discuss
that
it's
like
we
don't,
have
a
choice
right.
This
is
a
there's
only
one
when
you
have
a
fixed
exchange
rate,
you
have.
B
D
Control
over
you
months
or
policy,
and
you
just
have
to
put
it
at
the
spot,
Quinn
keep
you
hold
up
your
your
fixed
exchange
rate
and
again,
as
we
have
served,
there's
a
big
imbalance
in
how
the
balance
sheets
of
market
makers
look
right
now.
So
there
is
a
need
for,
for
us
likes
likes
significant
action
and
that's
what
we
think
a
2%
increase
represents
and
although
I
mean
so
we're
not
saying
it's
low
or
high
or
something,
but
we
definitely
they
could
be
as
too
high
and
we
hadn't
reduce
it.
D
It's
gonna
be
the
same
thing
like
it's
gonna,
be
all
about
us
teaching
the
community
and
us
teaching
each
other
kinda
discussing
like
I,
mean
the
other
foundation
right
in
the
and
the
foundation
risky
and
then
also
getting
feedback
from
the
community
and,
ultimately,
together
with
the
community.
Creating
this
like
scientific
process
that
we
used
to
set
up
what
our
monetary
policy
looks
like.
C
Yeah
I
just
like
to
add
I
just
like
to
add
to
that
that
I
would
recommend
thinking
about
about
rates
in
either
an
absolute
turns
so
like
we're,
increasing
it
by
two
percent
or
in
terms
relative
to
other
rates,
but
usually
people
don't
measure
rates
relative
to
to
themselves.
So,
for
example,
I,
don't
think
it's
very
meaningful
to
say
that
the
we've
increased
something
by
five
times.
It's
a
lot.
C
You
can
see
it
a
lot
clearer
if
you
compare,
if
you
compare
this
rate
to
other
rates,
so
like
I,
don't
know,
look
at
how
I'm
not
just
saying
to
do
that
to
feel
good
I'm,
just
saying
that
this
is
how
it
makes
sense
to
look
at
rates
in
a
macroeconomic
setting.
Compare
this
rate.
Whatever
you
think.
Inflation
is
compare
this
rate,
how
much
US
Treasury
bills
are
paying.
C
C
You
look
at
there's
the
spread
over
some
over
some
baseline
rate,
because
race
will
change
over
time
and
Basel,
but
the
thing
that
there's
always
two
components
that
govern
rates,
which
is
kind
of
the
the
macro
condition
and
then
the
local
condition
of
the
the
instrument
the
year
that
you're
talking
about
and
I
think.
This
is
also
the
right
way
of
looking
at
rates
here.
I
can
I
can
answer
this
question
of
from
Kluger
about
stability
theme,
multi
collateral,
yeah.
C
Yeah
sorry
so
clearly
asks
so
in
the
multi
collateral
world.
The
stability
fee
for
each
collateral
would
cover
two
things:
one
collateral
specific
risk
and
two
holding
the
peg
by
balancing
the
supply
and
demand
of
die.
How
will
that
work
in
practice?
Will
there
be
a
single
premium
that
has
added
to
all
CDP
types
or
will
the
rates
differ,
and
these
will
be
used
to
drive
a
better
distribution
of
collateral,
favoring
collateral?
We
are
under
bait
compared
to
an
ideal
risk
of
is
important
yeah.
This
is
it.
C
This
is
a
terrific
question,
so
I
think
there's
two
components
with
to
it.
One
is
you're
asking
so
technically
we'll
give
you
those
two
components:
well,
should
we
think
of
rates
as
consist
consisting
of
a
base
rate
and
a
spread
based
on
the
kind
of
base
on
the
risk
for
the
collateral
and
the
answer?
That
is
absolutely
I,
think
that
is
exactly
how
we
should.
How
would
you
think
of
rates
for
the
reasons
that
I
just
said?
That's
that's
how
everyone
always
thinks
of
rates.
C
The
second
question
is
technically
speak
or
howl
at
work
in
practice.
Will
there
be
a
single
premium
that
is
added
to
all
CVP
types,
I
mean
there's
also
a
technical
question
of
how
this
is
represented
in
the
system,
so
this
is
to
literally
store
a
variable
that
represents
the
kind
of
base
rate
and
then
each
each
collateral
type
has
a
has
a
spread
to
put
on
top
of
that
and
I
think.
C
B
Well,
the
first
point
to
to
speak
to
is
the
simple
aspect
of
where
the
actual
nominal
level
of
the
rate
is
right.
Now,
the
short
term
USD
LIBOR
curve
averages
are
to
run
about
two
and
a
half
percent,
so
it
is
roughly
a
good
guide
of
equal
is
thing
that
sort
up
a
return
profile
with
what
you
can
get
in
traditional
market,
which
then
will
then
talk
to
what
rune
had
previous
said
and
what
live
has
previously
said
about
not
addressing
the
certain
imbalances
coming
from
cheap
credit.
B
Now,
the
chief
credit
is
cheap
compared
to
what-
and
this
is
what
live
effectively
mentioned
right
now,
you
have
to
decide
what
it
is
cheap
compared
to
now.
The
simplest
thing
in
the
world
is
considering.
We
are
pegged
to
the
dollar,
have
a
look
at
the
USD
LIBOR
short-term
curve.
You
come
up
to
about
two
point.
Eight
three
percent
up
to
one
year
averages
around
about
two
point,
four
to
two
point:
five,
so
it's
just
equalization
over
there
and
it
helps
very
clearly
understanding
that
point
of
view.
B
Then
you
do
similarities
and
by
including
that
you
will
increase
the
diversification
benefit
of
your
your
your
collateral
portfolio.
Now
there
is
an
dynamics
here.
The
one
is
use
your
portfolio
to
look
after
the
system.
The
other
dynamic
is.
How
do
you
induce
folks
to
use
more
or
less
of
a
particular
collateral?
So
those
aspects
first
have
to
be
addressed
from
a
risk
point
of
view,
but
to
try
and
specify
how
one
particular
collateral
type
should
be
induced
over.
Another
is
a
constraining
factor
on
how
you
build
your
collateral
portfolio.
B
Overall,
it
can
be
considered,
can
be
done.
It
just
is
a
high
level
of
complication.
I,
don't
know
if
that's
answered
your
your
question
directly,
but
I
fitting
I
took
even
after
so
how
you
actually
look
at
this
from
a
high
level,
all
the
way
down
to
the
you
know
specific
collateral
type
as
well.
This
goes
to
another
very
important
point
that
live
brought
up,
and
that
is
the
supply
and
demand
imbalance
that
occurs
naturally
through
the
ebbs
and
flows
of
the
economy.
B
You
observe
it
because
of
extraneous
factors
that
come
in
affected
things
change,
and
you
have
to
react
to
the
perfectly
now,
there's
two
aspects
that
we
rely
on
the
first
one,
and
this
might
actually
talk
to
some
other
questions.
I've
seen
in
the
chat,
the
first
one
is
governing
dynamics
should
kick
in
your
arbitrage
process
should
kick
in.
But
the
remaining
question
is
what
centage
difference
from
the
peg.
Would
you
expect
all
these
governing
dynamics
to
kick
in
and
that's
where
you
actually
need
a
policy
tool
to
augment,
or
rather
support
those
governing
dynamics?
B
Because
if
you
ask
in
CDP
holders
to
realize
arbitrage
at
what
level
will
they
do
that?
Will
they
do
that
when
the
peg,
when
the
the
the
cost
of
BI
is
around
97
cents
to
the
dollar,
implying
a
potential
3%
a
brush
profit?
Well,
you
don't
know
because
the
fact
is
underlying
that
volatility
of
the
collateral
that
they
place
versus
the
the
friction
and
the
transactional
cost
of
doing
this.
B
But
that,
in
essence,
is
what
you
rely
on
that
governing
dynamic
and
if
that
governing
dynamic,
proves
to
create
too
much
of
a
marginal
or
a
marginal
difference
that
is
material
to
the
actual
peg.
You
need
a
policy
instrument
to
come
in
and
actually
address
it
appropriately,
and
the
first
steps
just
to
go
back
to
the
initial
point
is
the
way
that
you
create
a
policy
instrument
to
be
more
reflective
of
what
is
a
good
benchmark.
B
Interest
rate
is
by
merely
looking
at
the
the
the
costume
structure
surrounding
the
thing
you
pegging
against,
which
is
the
dollar.
So
therefore,
the
USD
LIBOR
curve
would
be
a
good
reference
point.
You
can
use
anything
else.
You
can
use
inflation.
You
can
use
constructs
that
you
see
on
the
blockchain,
but
considering
the
fact
that
we
are
effectively
borrowing
a
monetary
policy
from
the
from
the
US
by
pegging
it
may
be
a
very
good
first
protocol.
A
F
Sure
so
the
question
was
just:
can
you
guys
hear
me?
Okay
question
was
just
basically
you
know
after
multi
collateral
diet
launches.
Would
we
be
able
to
create
different
kinds
of
CDP's
for
the
same
asset
type
that
I
have
different
collateralization
ratios
and
stability
feeds
so
that
users
can
have
different
options?
For
example,
you
know
one
eats
EDP
type.
That
would
require
finessing
at
500%
collateralization
ratio
and
have
a
half
a
percent
interest
rate
while
another,
each
ccdb
type
could
have.
F
B
We
talking
about
the
the
token
specifically.
Are
we
addressing
the
risk
profiles
of
users,
because
I
mean,
if
you
got
like
seriously
risk-averse
people,
they
want
the
risk
profiles
that
you
know
tend
to
their
in
senses,
all
the
way
to
no
risk
loving,
and
you
know
those
guys
will
want
some
serious
CDP's
that
I
have
a
higher
probability
of
being
liquidated.
Is
that
more?
What
are
we
addressing
I'm.
F
More
just
talking
about
the
ability
to
have
you
know
two
different
types
of
CDP
on
the
same
asset
class.
That
kind
of
you
know,
because
I
can
sort
of
see
the
issue
right.
Like
a
question
that
was
asked
before
about
you
know,
is
there
gonna
be
a
single
premium
at
it
all
CDP
types
like
if
you
have
options
where
you
know
all
represent
clouds
action
ratio?
You
have
tons
of
these
lying
around
and
you
just
some.
You
know
cheap.
What
I
do?
F
B
Is
a
consumer
surplus
argument?
Basically,
it
really
is
a
case
of
saying
you
know
what
do
consumers
want
in
the
space
at
the
moment?
I
believe
we
are
addressing.
You
know
a
certain
aspect,
and
if
we
then
granular
eyes
that
aspect,
then
we
definitely
will
be
able
to
capture
a
lot
more
of
consumer
surplus
by
saying
sure
you
can
have,
let's,
let's
have
a
look
at
it
as
a
example
and
obviously
live
in
ruin.
B
Now
that
range
creates
like
a
surface
of
possibilities
because
it
attends
to
not
only
the
risk
profile
but
the
the
Tillet
e
function
of
each
user
and,
as
a
consequence,
by
implicitly
using
a
CDP
on
that
surface.
You've
creates
an
optimal
structure
for
yourself,
but
that
is
very
much
in
the
hands
of
the
users.
The
ability
to
provide
this
I,
don't
think,
is
terribly
difficult,
but
in
terms
of
understanding,
the
flexibility
around
this
problem
probably
will
require
a
certain
amount
of
time.
So
the
short
answer
is
sure:
I,
don't
think,
that's
a
problem.
B
The
the
medium
term
answer
is
I.
Think
that's
going
to
go,
neither
a
sort
of
a
learning
process
before
you
could
actually
offer
that
because
we
offer
something
that
has
certainly
high
risk
units
and
someone
isn't
aware
of
it
or
someone
is
not
aware
of
the
fact
that
they're
actually
selecting
one
specific
for
a
risk,
loving
incentivized
user.
You
might
have
an
asymmetry
information
which
may
prove
a
dislocated
function,
and
you
might
have
you
know,
problems
effectively.
B
C
I
think
the
only
thing
I'd
add
to
that
is
it
from
a
technical
point
of
view.
It's
it's
completely
possible
and
it's
part
it's
one
of
the
features.
That's
the
system
has.
Are
they?
The
only
whole
thing
about
it
is
the
is
understanding
user
needs
to
make
sure
that
this
is
deployed
in
a
way,
that
is
that
addresses
them
well,
but
I.
Think,
thankfully,
because
this
is
a
community
driven
project
where
the
users
are
actively
participating
in
governing
the
system.
C
I
expect
that
the
feedback
there
will
be
very
direct
and
it
will
be
actually
very
easy
to
determine
what
you
know.
What
spectrum
of
CEP
types
are
required
to
meet
different
different
use
cases
I,
don't
imagine
that
there
would
be.
You
know,
I,
don't
imagine
that
we
would
have
to
add
like
a
thousands
of
different
CDP
types,
because
we
don't
know
what
users
need.
I
think
that
there
will
probably
be
a
dialog
where
we
can
get
a
very
good
idea.
A
Live
while
we
have
you,
there's
a
question
that
I
want
to
address
before
the
end
of
the
meeting.
It
might
be
a
bit
of
a
curveball
for
you,
but
there
was
some
questions
in
the
initial
reddit
thread
about
how
this
stability
fee,
this
dynamic,
balancing
mechanism
that
we've
been
proposing,
how
that
syncs
up
with
the
mechanisms
that
were
meant
to
be
addressed
by
the
TR,
FM
and
I'm,
wondering
if
you
can
maybe
compare
and
contrast
those
two
solutions
for
us,
yeah.
C
Okay,
I
can
talk
about
that.
I
think
this
is.
This
is
a
very.
This
is
a
very
involved
question,
so
this
could
this
answer
could
go
on
for
hours,
but
the
crucial
points
are
these,
so
the
the
the
the
goal
of
adjusting
risk
parameters
such
as
the
stability
fee
is,
is
the
same
as
the
goal
of
TR
FM
was
so.
The
these
things
are
trying
to
solve.
C
Basically,
the
same
problem
and
I
would
say
that
this
perspective
of
adjusting
risk
parameters
rather
than
using
the
TR
FM
to
adjust
target
price
is
just
a
different,
is
a
different
kind
of
technical
way
of
tackling
the
same
problem.
So
so
these
these
are
I'm
doing
these.
These
things
are
closely
related.
I
can
comment
on
what
are
the
main
issues
that
the
jump-out,
if
you
consider
the
TR
FM
and
which
are
not,
which
we
want
to
avoid
with
with
what
we
do
so
one
serious
issue,
I
think
with
erm,
is
a
usability
issue.
C
People's
people
are
mostly
thinking
in
the
local
currency
or
the
US
dollar
or
wherever
they
are
adjusting
the
value
of
that
unit
of
account
as
a
as
a
stability
mechanism
is
very
confusing,
so
it
would,
it
would
create
a
lot
of
difficulties
for
almost
from
almost
all
you
know,
whatever
gaps
that
used
I
ever
use
case,
every
every
partner
that
the
will
do
we
have.
Every
user
would
find
this
to
be
an
inconvenience
if
they
had
to
keep
track
of.
What
is
the
die
to
USD
target
price
that
they
always
need
to
convert
them?
C
They
can
just
imagine,
for
example,
if
you
know
one
of
the
things
that
we
really
believe
in
is
people
using
Thai
as
a
quote
token,
on
decentralized
exchanges,
because
we
think
the
decentralize
exchanges
are
great
and
then
one
of
the
main
thing
that
they're
missing
is
just
being
able
to
trade
against
it.
Stricter
trade
against
a
stable
unit
of
account,
which
is
what
we're
we're
we're
create.
C
Imagine
what
it
would
be
like
if
there
was
all
these
centralized
exchanges
that
you
could
trade
against
die
but
die
was
like
you
know,
1.0
$2,
and
then
all
the
rates
were
also
when
you
were
comparing
on
in
a
different
window.
You
were
looking
at
what
was
going
on
at
gia,
X
or
something
it
would
all
be
slightly
off
and
not
to
have
a
calculator
out
to
do
all
these
conversions.
It's
obviously
creates
a
lot
of
friction
and
non
dimensional,
a
bunch
of
other
applications
where
this
is
obviously
the
problem.
C
So
and
then
always
people
are
borrowing
it's
the
same
effect.
So
we
don't
want
to
do
this
thing
anymore,
where
we
change
the
change,
the
the
value
of
a
token
as
a
way
of
as
a
way
of
creating
interest,
because
it's
it's
just
technically
so
confusing
that
you
know
you
would
really
blame
anyone
for
finding
that
to
be
a
problem.
The
second
thing
about
TRF
em
that
is
potentially
problematic,
is
that,
or
at
least
is
how
it
wasn't
described
or
understood,
even
though
I
don't
think.
C
This
is
like
an
inherent
problem
to
to
the
idea
is
having
a
very
having
this
very
simple
feed
reliant
mechanism
for
adjusting
the
for
adjusting
the
target
rate.
So
you
don't
remember,
target
rate
is
the
rate
at
which
the
target
price
changes
having
this
very
simple
mechanism
for
adjusting
that
continuously
adjusting
the
the
target
rate.
Algorithmically
has
some
obvious.
C
Kind
of
I
would
say
stability
issues
if
the
quality
of
the
data
that's
coming
in,
isn't
sufficiently
high.
So
this
this,
this
T
RFM
mechanism
is,
is
running
based
on
a
on
a
feed
for
a
value,
SD
price,
and
it
was
never
completely
cleared
up.
What
is
where
would
this
feed
be
coming
from
how
exactly
be
measuring
this,
and
it
turns
out
that
actually
TR
FM
is
very
sensitive
to
exactly
how
to
how
this
price
is
calculated,
so
I
mean
just
give
you
a
comparison.
C
Another
place
where
we
used
feeds
is
to
is
to
assess
safety
of
cdpas
right,
so
we
use
feeds
for
EQ
SD
and
in
the
future
for
any
other
collateral
to
check
whether
to
TP
a
safe
one.
Nice
thing
about
that
is
that
the
stability
of
a
system
is
not
finally
dependent
on
how
that
feed
is
calculated.
So
we
know
that
it
provided
the
feed
is
giving
you
approximately
something
with
something
correct
something
that
was
correct.
Maybe
20
minutes
ago
some
average
over
pulled
over
a
large
number
of
exchanges.
The
system
is
going
to
work.
C
Okay,
because
that's
gonna,
that's
gonna,
be
decent
enough
expression
for
the
price
to
decide
whether
or
not
a
CVP
is
safe
right.
So
we
there's
no
there's
no
way
of
like
by
very
slightly
manipulating
the
the
price
that's
fed
in
and
for
collateral
feeds,
there's
no
way
of
exploiting
the
system
or
making
it
unstable,
because
that
prices
needs
to
be
approximately
correct.
C
It's
much
more
of
a
of
a
kind
of
a
granade
than
a
good
gun
type
thing
so
basically
be
somewhere
in
the
radius
of
the
right
price,
while
the
price
feed
that
you'd
feed
into
TR
FM,
if
you
were
using
TR
FM,
actually
needs
to
have
very
nice
properties,
and
if
it's
you
know,
potentially,
if
it's
manipulated
very
slightly
or
if
it's
calculated
in
the
wrong
way,
there
could
be
vectors
in
the
stability.
Theoretically,
the
I
think
it's.
It
takes
some
time
to
explain.
C
These
types
of
things,
that's
not
the
only
example,
but
there's
this
kind
of
thing,
which
means
that
we
have
to
think
more
deeply
about
how
to
set
rates
than
using
a
very
simple
formula
like
the
T
RFM
ones,
but
but
fundamentally,
I
very
much
believe
in
using
this
kind
of
approach
for
setting
rates,
and
this
is
exactly
what
we're
researching
at
the
moment.
That's.
A
For
us
know,
if
the
answered
some
of
my
questions,
I've
always
had
some
concerns
about
the
sort
of
a
volatility
feedback
kind
of
losing
control
in
the
TRF
model,
but
I
have
to
admit
that
I
only
vaguely
understand
it
in
the
first
place.
Just
do
we
have
any
additional
questions
in
the
chat
that
we
should
address
before
we're
over
time.
E
D
A
D
Guess
you
don't
really
know
how
much
it
is,
but
it's
definitely
like
we're.
You
know
the
more
precise
and
perfectly
we
set
the
interest
rates,
potentially
the
more
liquidity
we
get.
I
think
it
front
like
I
mean
it's
probably
something
like
we
want
to
get
it
close
enough.
You
got
to
be
pretty
much
hit
this
theoretical
maximum,
but
there's
definitely
a
look
like
it's
gonna.
Take
a
long
time
to
to
get
to
that
point,
and
it's
and
our
goal
is
to
make
it
like
an
iterative
process.
D
We
have
it
together
with
community,
basically
trying
to
sit
down
and
like
solve
this
problem
scientifically
over
the
next
couple
years
and
step
one
unlike
the
status
quo.
What
we're
doing
right
now
is
like
we
are
making
these
like
big
changes
that
are
coming
out
of
the
wrist
team
and-
and
you
know
we
we
basically
make
them
and
we,
you
know
they'll,
be
like
we
make
some
research
first
decide
that
they're
needed
you
can
post
them.
D
We
vote
them
in
they
come
through,
we
analyze
the
result
of
them,
and
then
we
either
do
nothing
or
we
do
another
change,
and
that's
actually,
how
like
I
mean
that's
how
they
you're
like
anything
like
this,
actually
really
works
right
like
how
the
system
fundamentally
needs
of
you.
Son
come
on
just
like
it's
just
like.
If
we
get
through
it
in
higher
speed
and
at
more
precision,
it
directly
improves
the
power
of
the
system.
So
that's
going
to
be
something.
That's
really
interesting.
A
A
No
well,
okay,
that's
great.
We
had
some
amazing
conversations
today
and
happily,
there's
even
more
cool
things
happening
in
our
chat.
So
if
there's
discussions
that
we
didn't
get
to
or
interesting
ideas
that
were
raised,
that
we
didn't
talk
about,
please
feel
free
to
take
them
over
to
Reddit
or
bring
them
into
a
rocket
chat.
Give
me
the
conversation
thanks.