►
From YouTube: Governance and Risk Meeting: Ep. 7
Description
In this Governance and Risk meeting, Steven Becker, President and COO of MakerDAO, will continue guiding us through his Governance Risk Framework document.
This is one of many governance-specific meetings we will be holding in preparation for the public launch of Decentralised Governance.
Governance and Risk Framework - Part 1: https://medium.com/makerdao/makerdao-governance-risk-framework-38625f514101
Governance and Risk Framework - Part 2: https://medium.com/makerdao/makerdao-governance-risk-framework-fc8939f3d4e9
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
And
welcome
to
our
August
23rd
episode
of
the
scientific
governance
and
risk
meetings
we've
been
having
with
Stephen
Becker
our
new
president
and
CEO
Oh
generally
I've
been
giving
some
introductions,
but
I
think
that
we're
so
far
in
the
weeds
or
at
being
sort
of
a
recap,
but
we're
so
far
in
the
weeds.
Now
that
that
might
not
be
a
useful
endeavor
and
angle,
let's
even
take
care
of
the
recaps
for
us
so
see
when
you
want
to
kick
it
off
for
us
cool.
B
So,
to
start
off
just
to
get
a
sense
of
what
we
previously
covered,
we
covered
the
risks
inherent
in
generating
and
I
using
CDP's,
not
just
to
recall
what
were
they
qualitative
risk,
exposure,
liquidity,
volatility,
correlation
and
price
theorist.
Now
the
reason
I'd
like
to
repeat
these
is
because
the
risk
parameters
that
we
are
currently
discussing
and
we'll
get
into
further,
especially
today,
actually
encapsulates
two
or
three
of
these
particular
risks.
So
the
one
aspect
that
we
actually
did
cover
in
depth
was
the
qualitative
risk
component
and
the
debt
ceiling
component.
B
But
for
today
what
I'd
like
to
do
is
go
over
the
the
debt
ceiling
component
again,
just
to
make
sure
that
we
have
a
good
sense
of
clarity
of
what
it
is
and
what
we're
trying
to
achieve
with
it
and
then
possibly
get
into
the
liquidation
ratio
and
sort
of
explore
the
extents
of
what
that
parameter
is
trying
to
control
for
and
how
we
can
go
about
actually
calculating
that
ratio
and
therefore
controlling
the
risk.
So
what
would
we
know
about
the
debt
ceiling?
Is
that
as
a
start
off,
is
that
this
is
a
parameter?
B
That
is
maybe
we
use
to
control
for
exposure,
risk
liquidity
risk
and
actually
contributes
towards
correlation
risk,
we're
in
a
correlation
risk.
We
will
have
a
look
at
maybe
in
another
core
I,
don't
think
we'd
be
able
to
get
to
that
today.
But
the
one
aspect
that
we
need
to
start
with
is
having
a
look
at
what
the
debt
ceiling
is
briefly
and
how
it's
actually
broken
down.
The
first
part
of
the
debt
ceiling
that
we
need
to
look
at
is
the
theoretical
debt
ceiling.
B
Now
that,
basically,
is
a
top-down
approach
in
understanding
what
that
means.
What
we
saying
is
that
the
trading
environment
is
particularly
liquid
and
we
think
we
can
actually
achieve
a
lot
in
terms
of
getting
rid
of
big
positions,
but
ultimately,
how
big
must
at
position
be
before
we
start
actually
impact
in
the
market
to
such
an
extent
that
we
cause
our
own
liquidity
problems,
so
the
debt
ceiling
as
a
parameter
has
this
theoretical
debt
ceiling
in
place,
which
is
from
a
top-down
approach,
is
just
a
percentage
of
the
total
tokens
that
are
issued.
B
B
So
there's
a
dual
role
over
here
that
the
practical
debt
ceiling
tries
to
cover,
and
once
we
figure
out
what
that
practical
debt
ceiling
is,
we
have
a
much
more
substantial
sense
of
what
we
can
be
exposed
to
if
the
entire
position
of
that
particular
token
had
to
be
liquidated
of
a
given
period.
So
the
first
one
was
theoretical
debt
ceiling,
which
said
you
know.
We
all
watched
the
liquidity.
What
is
the?
What
is
the
impact
we
would
have
on
that
liquidity?
B
The
second
one
says:
okay,
let's
start
considering,
both
liquidity
is
the
one
that
is
inherent
in
the
market
and
the
one
that
we
will
create
by
trying
to
liquidate
this
position
now.
The
third
component
to
the
debt
ceiling
is
more
the
defensive
debt
ceiling,
and
that
helps
us
with
couple
of
aspects.
B
In
a
particular
token,
you
may
find
that
some
irrational
exuberance
may
cause
people
to
climb
into
a
particularly
token
to
try
and
get
margin,
exposure
and
and
they're
very
a
better
exposure
to
the
irrational
exuberance
that
is
underlying.
That
particular
token.
We've
got
to
defend
against
that,
because
it
will
pose
a
risk
on
the
underlying
stability
of
the
system
as
well.
A
To
actually
jump
in
because
I
can
and
I
haven't
microphone.
So
what
if
it
joins
the
question
the
second
year,
so
the
debt
ceilings
from
a
very
high-level
perspective
kind
of
represents
the
the
portfolio
of
the.
How
in
how
deep
we
are
into
each
one
of
our
asset
classes,
which
represents
sort
of
a
report
from
a
user
perspective,
their
portfolio
of
assessments,
if
we
adopt
a
new
asset
class
fairly
large
asset
class,
for
example.
B
There's
a
positive
aspect
to
it
and
a
negative
aspect.
The
positive
aspect
is
that,
if
that
asset
was
large
enough
in
ubiquitous
happened
had
the
liquidity,
we
could
actually
incorporate
that
into
our
portfolio
with
a
larger
than
normal
defensive
debt
ceiling.
So
let's
say,
for
example,
now
we
start
everything
off
at
ten
million
dollars
each
and
we
slowly
grow
those
ten
million
dollars
each
as
we,
you
know,
facilitate
the
growth
of
this
of
the
system.
B
We
find
that
there's
a
silver
back
token
that
is
incredibly
liquid
and
is
fantastic,
and
you
would
love
to
have
that
on
board.
If
the
portfolio
allows
for
it,
we
don't
have
to
start
at
ten.
We
could
start
at
50
because
it
would
be
in
relation
to
the
portfolio
as
a
whole.
It
would
have
a
wonderful
I
would
say
negative
correlation
effect,
which
would
add
to
the
stability
of
the
portfolio.
But
that
being
said,
the
downside
is
that
it
is
very
much
dependent
on
where
the
collateral
portfolio
is
at
the
moment.
B
If
you've
got
massive
exposure
to
just
or
rather
a
concentrated
exposure
to
ether
and
you've
got
this
wonderful
asset.
That's
going
to
come
into
play,
we're
going
to
make
sure
that
that
asset
doesn't
include
itself
and
just
become
a
new
concentration
problem
we
have
so
it
is
always
with
respect
to
what
the
collateral
portfolio
would
look
at
would
look
like
again.
We've
got
to
keep
in
mind
that
our
objective
is
the
integrity
and
stability
of
the
system,
not
to
grow
an
investment
portfolio.
B
So
to
that
extent
we
need
to
use
the
inherent
risk
features
that
we
have
created
for
ourselves.
Like
the
diversification
benefit,
we
need
to
use
that
as
capital
buffers
to
make
decisions
with
respect
to
which
collateral
types
we
need
and
at
what
time,
it's
not
to
say
that
we're
going
to
dismiss
lateral
types
that
wish
to
be
a
part
of
us
it'll
just
be
more
case.
B
Well,
if
you
are,
you
know
very
significance
in
terms
of
your
trade
representation
in
terms
of
your
size
that
will
be
taken
into
consideration,
but
if
not,
you
might
have
a
very
small
debt
ceiling
and
that
would
grow
exponentially
as
the
whole
system
and
has
hold
collateral
portfolio
grows
as
well.
Once
again,
this
is
not
an
investment
portfolio.
We
don't
have
the
ability
to
control
uniformly
how
the
collateral
as
a
broad
representation
will
grow
in
the
portfolio
because
only
set
limits
and
hopefully
find
that
all
those
portfolios.
B
Oh
sorry,
all
those
collateral
types
will
grow
in
unison
together,
and
then
we
can
increase
the
debt
ceilings
for
each
of
them
and
then
see
them
grow
as
well,
but
that
is
highly
unlikely,
you're
going
to
see
concentration
on
one
grow
and
that
might
have
to
go
up
by
itself
where
others
might
may
take
a
little
bit
of
time.
So
the
kind
of
two-sided
answer
to
your
question
is
yes,
you
would
basically
adjust
for
bad
collateral
type.
B
You
wouldn't
address
anything
else,
and
if
that
collateral
type
was
too
concentrated,
we
may
have
to
just
keep
it
a
little
bit
small
constrained
until
the
rest
of
the
portfolio
catches
up,
but
it
doesn't
take
away.
The
negative
side
is
that
it
doesn't
take
away
the
fact
that
we
would
not
include
it
if
it
is
a
quality
asset.
It
has
to
be
included.
A
B
B
In
this
case,
emergency
shutdown
is
exactly
what
it
says.
It
stops
the
whole
system,
but
in
terms
of
a
normal
operating
system
when
CDP's
get
boots,
you've
got
to
consider
the
fact
that
CDP's,
maybe
but
randomly
because
each
person
will
have
a
different
risk
profile
and
they'll
have
different
collateralization
to
debt
ratios.
B
But
I
am
of
the
opinion
that
if
you
just
take
a
very
conservative
view
from
a
dead
ceiling
point
of
view
and
just
assume
that
it
was
one
very
large
CDP
and
it
all
got
bitten
at
the
same
time
in
the
normal
course
of
operations.
How
would
that
affect
the
unlined
market
in
which
you
want
to
actually
divest
yourself
of
that
particular
asset?
B
These
are
basic
thought
experiments,
but
in
those
thought,
experiments,
application
of
which
builds
in
natural
buffers
for
ourselves,
not
considering
the
fact
that
people
have
different
incentives,
has
a
diversification
benefit
in
there
as
well
and
have
built
in
because
not
everyone
is
going
to
wait
for
or
not.
Everyone
is
going
to
go
and
purge
collateral
at
165
percent,
with
with
respect
to
eat.
As
an
example,
you
can
have
a
distribution
of
CDP
users
can
have
a
distribution
of
of
pi
users.
You
can
have
a
distribution
of
keepers.
B
C
Yeah,
so
my
question
is
what,
if
the
defensive
like
again
like
to
add
to
your
point
that
you
mentioned
above
before,
what
is
the
defensive
debt
ceiling
that
we
have
calculated
is
higher
than
the
other
two
debt
ceilings
like
the
theoretical
one
and
the
other
one.
So
would
it
mean
like
because
for
us,
like,
diversification,
is
more
important
with
the
defensive
debt
ceiling,
take
take
precedence.
B
In
terms
of
subsets,
so
the
theoretical
debt
ceiling
is
the
biggest
component
right.
Okay,
within
that
the
subsets
is
the
practical
debt
ceiling
and
in
practical
debt
ceiling,
the
subset
is
the
defensive
debt
ceiling,
so
the
defensive
has
to
basically
try
and
grow
into
the
practical,
and
when
it's
reached
the
practical,
that's
where
it
stops
and
it
can
only
then
the
practical
can
grow
into
the
theoretical,
given
that
the
online
liquidity
becomes
more
prevalent.
So
there's
this
theoretical
aspect
within
it,
the
practical
aspect
and
then
with
energy,
but
be
the
defensive
aspect.
C
B
One
aspect-
and
this
is
probably
for
a
much
broader
discussion,
but
emergency
shutdown
can
occur
for
a
number
of
reasons.
You
could
have
actually
everything
working
absolutely
perfectly
nothing
going
wrong,
but
in
one
moment
you
might
find
somehow
a
bad
actor
has
got
involved
and
is
trying
to
influence
the
system
in
an
untoward
manner.
Now,
given
the
fact
that
nothing
is
wrong,
you
wouldn't
have
all
the
levers
in
a
triggering
flags,
but
it
will
become
evident
to
the
the
global
settlers
that
something
is
a
mess.
Something
is
wrong:
maybe
it's
someone
influencing
the
price.
B
Maybe
it's
someone
trying
to.
Basically,
there
is
there's
a
lot
of
permutations
so
trying
to
focus
emergency
shutdown
on
a
particular
collateral
type
or
the
collateral
portfolio
implies
that
there
was
some
serious
bad
acts
of
provisioning
going
on,
and
if
that
is
the
case,
then
you
would
get
emergency
shutdown
if
you
would
have
like
a
complete
and
utter
Black,
Swan
event
happening.
You
can't
think
about
Black
Swan
events
in
a
traditional
sense
where
it's
like
this
massive
correction,
and
we
hope
that
you
know
it
will
eventually
come
back
through
policy
mechanism.
B
In
this
extent,
the
be
the
Black
Swan
would
be
something
quite
colossal.
It
might
be
more
aimed
at
the
ethereum
blockchain,
just
stopping
for
some
reason:
I,
don't
know
how
there
would
happen.
I
have
no
idea,
but
that
kind
of
extent
would
be
more
systematic
than
it
would
actually
be
specific
to
the
collateral
portfolio.
All
the
you
know
the
collateral
types
now.
A
B
Debt
ceiling
of
10
million-
you
know
you
look
at
and
go
well
that
sucks.
We
need
to
basically,
you
know,
adjust
for
this
quickly,
but
in
terms
of
an
emergency
shutdown.
For
that
basis,
one
needs
to
take
that
into
consideration,
wait
and
say
well.
This
is
not
the
existential
threat.
This
is
just
messing
with
one
component
and
we
need
to
basically
drive
home
how
we
can
fix
that
particular
component.
I.
Think.
A
So
the
question
is
outside
of
this
specific
example
that
we're
talking
about
right
now,
but
in
general,
is
it
possible
for
us
to
shut
down.
So
if
we
had
a
silver
token,
we
realize,
if
something
think
he
was
happening
there,
what
is
it
possible
to
perform
an
emergency
shutdown
on
that
specific
asset
class,
or
is
there
a
different
term
we'd
use,
for
that
would
be
just
simply
freeze
all
trading
of
or
all
CDP
times
it
occurred
to
me
that
we
don't
I.
B
Would
imagine
that
there
is
a
capability
to
do
it?
I?
Imagine
that
there
is
a
way
that
we
could
do
it,
but
basically
it
boils
down
to
an
article
being
influenced
or
a
collection,
a
collection
of
articles
being
influenced,
and
we
do
have
security
mechanisms
in
place
to
know
check
the
pricing
before
it
affects
us.
B
Well,
if
you
use
the
silver,
if
silver
does
correct
by
70
percent
in
one
day
and
you
can
see
in
the
traditional
market
becomes
very
clear
and
evident
that
this
is
the
the
correction
value,
then
that
is
the
way
the
market
is
operating
and
CDP's.
We
will
be
bitten
and
most
probably
a
large
chunk
of
the
CDP's
will
be
bitten,
and
we
need
to
make
sure
that
the.
B
A
It's
interesting
a
point:
that's
a
good
observation,
so
there's
there's
a
difference
between
an
asset
class
just
simply
failing
in
a
sort
of
a
normal
market
manner,
as
opposed
to
some
group
of
bad
actors
attempting
to
manipulate
the
system.
So
if
silver
does
correct
and
that's
we've,
essentially
people
have
paid
their
money,
they
took
their
chances
and
then
those
CDP's
will
simply
fail,
and
if
the
system
is
designed
correctly
it
won't.
It
won't
prevent
them
present
an
existent
existential
threat
to
the
rest
of
the
system.
Anyways,
where.
B
The
way
I
understand
this
is
that
you
have
to
have
a
we
operate
in
an
environment,
and
if
that
environment
has
a
systemic
shutdown,
then
it
affects
us
and
that
environment
at
the
moment
is
the
etherion
blockchain.
Some
reason
that
thing
shuts
off
something
goes
wrong
with
it.
Then
you
know
by
default.
So
do
we,
but
in
terms
of
us
as
a
system,
you
basically
have
to
identify
the
permutations
of
bad
actors
and
either
list
them.
B
B
Basically,
there's
two
ways:
there's
obviously
the
one.
The
one
thing
that
seems
to
be
a
natural
understanding
and
I
need
to.
Basically,
you
know
adjust
that
is
that
debt
ceilings
don't
just
grow.
You
may
have
to
the
the
trading
environment
may
deteriorate.
The
the
fundamental
qualities
of
the
actual
organization
may
pivot.
These
are
the
things
that
will
fundamentally
change
the
way
the
risk
teams
will
perceive
these
organizations
and
its
opens
I
mean,
for
instance,
let's.
Let's
imagine
that
we
used
Apple
stock
as
collateral
and
then
tomorrow,
Apple
decided.
They
don't
want
to.
B
You
know,
sell
devices
anymore.
They
want
to
get
into
laundry
service,
I
mean
Apple
will
absolutely
just
tank.
Everything
would
just
change,
but
the
point
being
is
because
everyone
is
constantly
keeping
an
eye
on
Apple.
The
fact
that
they
want
to
turn
into
a
laundry
service
company
would
have
probably
been
seen
a
long
time
ago.
You
would
have
seen
it
in
the
fact
that
there's
there
their
profits
were
coming
down,
you
could
see
it
in
the
fact
that
their
devices
weren't
selling-
you
could
see
it
in
the
change
of
management.
B
These
things
would
slowly
give
you
the
the
flags
to
start
saying.
Qualitative
assessment
is
dropping.
We
may
have
to
even
consider
excluding
this
these
guys
from
the
portfolio
in
total.
Further
to
that,
you
may
even
address
the
fact
that
the
trading
environment
has
changed.
Is
that
one
of
the
the
basic
aspects
of
understanding
when
a
potential
correctional
crash
comes
along
is
having
to
look
at
the
behavior
of
the
market
just
before
it
actually
occurs.
There
is
a
sense
of
erratic
behavior
where
things
become
very
hectic.
B
Folks,
do
not
understand
whether
they're
coming
or
going
it
doesn't
mean
that
it
is
to
foretell
the
future,
but
it
generally
gives
you
a
good
indication
and
then,
from
there
risk
teams
could
pick
up
the
change
in
the
nature
of
the
the
trading
environment,
which
should
affect
the
underlying
figures,
the
calculations
which
then
should
start
bringing
down
the
dead
ceilings.
If
you
have
a
binary
event,
if
you
have
a
binary
events,
that
I'll
concede
is
a
lot
harder.
B
So
what
happens
so
David
goes
on
to
say
what
happens
when
you
need
to
lower
the
debt
ceiling,
but
there's
a
maximum
amount
of
diatone
for
that
as
already
well.
It
just
happened
that
the
extra
die
of
the
debts
well,
the
the
thing
about
the
debt
ceiling
is
that
obviously
you
are
going
to
have
to
first
message
this
out
and
say:
listen.
We
need
to
drop
the.
B
We
need
to
drop
the
debts
here
in
terms
of
actually
dealing
with
something
that
is
frantic
and
hectic
like
dropping
the
debt
ceiling,
let's
say
from
20
down
to
down
to
10,
but
you
know
there's
already
18
out.
So
that
means
for
some
reason,
you've
got
8.
You've
got
8
million
die
out
there.
That
needs
to
come
back
well
either.
You
can
do
it
in
a
very
progressive
format
and
transition
slowly
or
if
there's
something
drastic
it
just
pretty.
Much
would
have
to
be
a
case
of
well.
No
more
can
be
taken
out.
B
We
have
to
actually
cap
it
at
the
18,
but
the
effective
amount
will
be
10
and
then
folks
will
start,
hopefully
bringing
back
their
they're
dying,
closing
out
their
CDP's
bad
events.
I
can't
I
can't
give
you
the
the
absolute
details
or
what
happens
in
that
event,
but
I
mean
it's
a
very
good
point.
So.
A
I
think
the
assumption
would
be
that
if
we
had
a
Tesla
back
asset
class
and
then
somebody
was
watching
Tesla
tweets,
they
would
they
would
probably
Elon
Musk
tweets.
They
probably
get
a
sense
that
something
was
not
great
in
the
world
of
Tesla
and
they
would
start
you
know
down
there.
Cdp
positions
that
came
back
died
right.
So
the
expectation
is
the
people
that
opened
that
class
of
CDP
s
would
be
watching
the
Oracles
and
watching
their
own
ecosystem
and
trying
to
get
out
of
their
position.
I.
B
Mean
there
I
mean
Davis:
question
is
spot
on
I
mean
if
you
do.
There
are
certain
reactive
components
that
we
have
in
place,
but
in
terms
of
all
the
possible
permutation
of
events
out
there,
you
cannot
you
try
to
cover
for
as
much
as
possible,
but
this
you
know
the
thing
about
a
Black
Swan.
Everyone
talks
about
it
as
a
thought
experiment,
but
way
that
it
manifests
itself
is
never
in
a
way.
You
expect
I
mean
that's
why?
B
Long-Term
capital
management
system
and
its
demise,
you
know
they
talk
in
hindsight's
about
how
too
much
risk
was
taken
on
the
leverage
was
almost
infinite
and
so
on
and
so
forth,
and
you
think
well,
you
know
you
are
talking
about
nobel-winning.
Why
is
winning
you
know
mathematicians
and
economists?
It's
not
like.
They
didn't
consider
that
it's
just
that
when
you
know
Russia
defaulted
and
it
took
them
into
a
tailspin,
not
just
because
of
the
default,
but
because
of
what
their
counterparties
did
as
a
consequence
of
it.
B
So
these
things
didn't
just
come
from
like
a
miscalculation
or
Mis
assessment.
It
comes
from
the
overarching
fact
and
when
something
goes
wrong,
the
the
domino
effect
that
you
get
from
that
is
not
generally
is
never
going
to
be
something
you
anticipate
so
you've
got
to.
Basically
do
this
thing
called
stress
testing,
where
you
think
okay.
B
Well,
if
the
worst
case
I've
ever
seen
is
a
10%
drop
in
one
day,
then
you
have
to
start
with
the
words
like:
let's
assume
that
it's
15
plants
and
not
10,
and
the
problem
with
that
is
that
you
go
to
your
your
board
of
directors
and
you
put
forward
a
risk
stress
assessment
at
15%.
They
can
ask
you:
where
did
you
get
that
figure
from,
and
you
say
well,
it
is
to
basically
encapsulate
those
risks.
B
We
can't
even
think
about,
because
we
have
no
idea
how
they're
gonna
look
and
we
may
have
an
order
of
magnitude
that
is
so
low
that
we
don't
know.
So
the
point
being
is
to
make
sure
that
your
exposure
to
all
these
possibilities
is
constrained
into
a
world
where
you
think,
okay,
what
is
what
can
you
do
from
a
rigorous
point
of
view?
What
can
you
do
from
an
economic
capital
20
view
and
then
there's
just
stuff
that
you
cannot
possibly
even
fathom
but
because
you've
kept
your
exposure
well
contained
and
well
controlled?
B
If
you
do
have
an
absolute,
you
know,
cataclysmic
events
that
you
can.
You
know
pigeonhole
hundred
black
swan.
It's
just
that
you
are
more
likely
to
come
out
of
it
on
the
other
side
that
is
effectively
what
risk
management
does.
From
that
absolute
point
of
view,
it
just
increases
your
chances
of
survival.
I.
A
B
B
I,
remember
the
first
time
I
joined
an
investment
bank,
I
was
sitting
around
a
boardroom
table
and
the
head
of
operations
posed
a
question
to
everyone
saying
what
is
the
most
important
risk
that
you
have
facing
a
Trading
Bank
other
than
operations
and
everybody
was
going
on
followed,
silat
CEO
this
or
that,
and
it
became
she
just
looked
at
everyone
and
said
no,
it's
liquidity.
Liquidity
is
the
underlying
defining
factor
of
of
your
ability
to
trade
and.
B
Liquidity
assessments
on
a
continuous
basis
is
really
the
thing
that
you've
got
to
keep
your
home
at
all
times,
because
you've
got
to
assume
that
when
you
create
your
debt,
very
much
like
when
you
create
your
liquidation
ratio,
that
liquidity
that
you
see
will
persist
for
a
given
period
of
time.
It
is
unfortunate
you
can't
you
can't
decide
that
everything
works
really
well
now,
and
then
we
assume
tomorrow
it
all
disappears
because
that's
the
case,
then
those
are
bad
assumptions
and
we
shouldn't
be
doing
business
in
that
environment.
B
You
have
to
basically
make
critical
assessments
to
saying
what
you
see
now
can't
assist
for
at
least
the
next
three
months
so
assume
in
the
liquidity
provided
so
the
quiddity
profile
that
you
see
there
actually
has
it
persistence.
Then
you
can
do
calculations
of
the
back
of
it.
Some
of
the
stuff
seems
a
little
sort
of
vague,
but
ultimately
you
must
remember
that
we're
talking
about
participants
in
market
human
beings
participating
in
the
market,
there's
a
group
psychological
dynamic
that
ensures
people
gravitate
towards
a
common
cause,
which
is
in
incentivizes
them
to
trade.
B
So,
ultimately,
if
that
incentive
is
strong
enough
and
they're
gravitation
is
strong
enough,
your
liquidity
will
come
as
a
as
a
consequence.
If
you
note
that
during
the
2007
eight
crisis,
the
the
biggest
concern
wasn't
the
fact
that
there
was
a
massive
credit
crash,
it
was
the
fact
that
the
the
banks
decided
that
it
did
not
did
not
want
to
provide
any
liquidity
whatsoever.
B
They
didn't
want
to
lend
it
out
because
their
incentives
were
gone,
and
that
is
the
underlying
idea
that
you
want
to
keep
in
the
back
of
your
mind
when
you
think
about
liquidity,
is
the
incentive
always
there
to
keep
trading?
What
is
causing
people
to
gravitate
towards
wanting
to
trade,
and
that
pretty
much
is
the
very,
very
premise
of
majority
of
where
all
your
risks
go
wrong?
It's
when
you
don't
consider
the
quiddity,
so
that
being
said
and
see,
are
there
any
other
questions
before
he
liquidation
ratio,
yeah.
B
You
would,
as
I
previously
mentioned
it,
is
you
have
to
make
an
assumption
that
the
liquidity
profile
that
you
see
today
will
persist
for
a
certain
period
of
time.
You
can't
assume
that
it's
going
to
be
exactly
the
same
next
year,
but
you
can
make
the
assumption
that
it's
going
to
be
pretty
much
the
same
over
the
next
three
months.
B
Obviously
it
is
a
statistic,
so
you'll
have
confidence
intervals.
You'll
have
all
these
metrics.
Your
bootstrapping
will
tell
you
that
how
far
your
your
your
metrics
may
be
off,
but
the
point
being
is
that
the
unlined
assumption
is
your
liquidity
profile
will
persist
for
the
next
three
months
in
the
crypto
space
three
months,
I
reckon
is
about
the
longest.
You
can
push
it
in
the
traditional
space.
Most
people
go
for
more
like
six
months
to
a
year.
You
know
looking
at
risk
management
and
assuming
the
the
mobility
profile.
B
What
is
interesting
on
that
basis
is
that
in
the
traditional
sense,
when
you
look
at
liquidity,
its
regime
dependent
you
you've
met,
you
may
have
a
current
regime
that
is
bullish
like
in
where
quantitative
easing
came
into
existence
in
2009
and
qe2.
Again
it
bolstered
the
whole
markets.
The
regime
turned
from
a
high
volatility,
a
bearish
market
into
this
low
volatility
sponsored
bull
market,
so
the
regime
shifted.
The
volatility
profiles
changed
so
that
aspect
has
a
you
know,
a
cyclic
nature
which
folks
try
and
keep
into
consideration.
B
Crypto
hasn't
been
around
long
enough
to
actually
start
formulating
these
proper
cycles
to
to
expressly
find
definition
in
cycles
because
the
economy
hasn't
grown
as
such.
But
given
that
you
need
to
be
cognizant
of
the
fact
that
you
need
to
express
some
sort
of
confidence
in
how
this,
especially
the
profile,
will
persist
over
a
certain
period.
Maybe
three
months
is
too
long.
Maybe
it
should
be
monthly,
but
from
a
risk
team
point
of
view,
the
MK
autocrat
holders
need
to
mandate
the
risk
teams
to
be
cognizant
of
it
every
day,
if
not
in
real
time.
B
Okay,
so
with
that,
what's
our
time,
timing
like
we've
got
15
minutes
so
I'm
gonna
I'm
gonna
spend
five
more
minutes
on
I'm
just
going
through
the
start
of
the
liquidation
ratio,
and
then
we
can
open
up
for
any
more
questions
or
dialogue
with
people
wish
I
can
carry
on
just
speaking
about
liquidation
ratio,
so
the
nation
ratio
is
a
parameter
that
encapsulates
two
risks:
volatility
and
liquidity
risk.
Now
the
liquidity
risk
assessment
is
a
little
different
to
that
which
we
do
for
the
debt
ceiling.
B
The
debt
ceiling
is
more
concerned
about,
you
know
the
size
of
the
ceiling
and
how
that
affects
the
actual
underlying
liquidity
profile
of
the
stock.
Here,
we're
looking
at
the
liquidity
assessment
from
the
bottom
up,
we're
basically
saying
that
if
I'm
hold
a
CD,
P
and
I
wish
to
liquidate
sorry,
the
CDP
gets
bitten
and
you
take
the
it.
You
basically
buy
the
collateral
and
you
wish
to
liquidate
it.
B
You
are
more
worried
about
the
liquidity
profile
that
you're
about
to
trade
in
to
where
the
debt
ceiling
you
more
worried
about
liquidity
profile,
bet
you
by
the
trade
into
and
impact,
so
the
liquidation
ratio
just
says.
Well,
let's
figure
out
what
the
liquidity
profile
looks
like
and
then
consider
the
next
issue,
which
is
volatility.
B
So
let's
chat
about
the
volatility
component
for
a
second
given
a
specific
point
in
time
like
today,
what
is
the
likelihood
of
token
ABC
dropping
by
one
percent
in
one
day
now?
That
statement
is
very
important
because
it's
made
up
of
two
components.
The
first
is
the
extent
in
terms
of
value
that
being
the
1%
and
then
the
second
is
the
time
period.
B
Now
that
encapsulates
the
full
extent
of
a
risk
expression
so
that
you
can
find
an
answer
for
it,
because
there
are
two
ways
that
you
need
to
look
at
it
either
you
worried
about
the
percentage
move
and
the
time
period
to
get
a
sense
of
the
probability
of
it
happening
all.
You
want
to
know
the
probability
of
something
happening
over
a
time
period
to
figure
out
what
the
extent
of
that
may
be.
Does
that
make
sense
before
I
carry
on?
Does
everyone
I
think.
A
B
Not
use
the
phrase
technical
analysis,
because
that
is
generally
something
that
is
is
kind
of
misconstrued.
It's
it's
a
set
of
tools
that
is
used
mostly
for
trading,
but
it's
generally
related
to
like
throwing
bones,
or
you
know
some
sort
of
mystical
thing
and
I
I,
don't
believe
in
that's
that
interpretation,
I.
Think
technical
analysis
has
first
principles
that
people
are
ignoring
in
terms
of
the
trading
environment
but
I
digress.
B
What
you
do
now
is
to
the
first
points
when
you
say
you've,
given
it
you've
got
a
percentage
move.
What's
the
probability
of
it
being
realized
over
a
specific
time
period
is
a
question
that
you
have
to
relate
to
the
underlying
distribution
of
the
returns
of
that
particular
asset,
and
so
what
you
do
is
take
a
certain
time
period
generally
more
than
let's
say
one
year.
It
depends
on
the
regime,
but
that's
a
another
subject
altogether.
B
Let's
say
you
take
the
returns
of
one
year
and
you
map
the
distribution
on
to
it
either
you
fits
a
distribution
to
it.
You
say
well,
this
is
the
the
the
looking
feel
of
the
distribution.
If
we
had
to
go
and
put
a
normal
distribution,
a
Gaussian
distribution
over,
what
would
this
parameters
look
like
I
mean
you
might
get
a
your
normal
distribution
is
defined
by
two
parameters,
the
the
mean
and
standard
deviation.
B
You
might
find
the
mean,
is
plus
point
five
percent
and
the
volatility
is
one
or
two
percent,
and
that
defines
your
your
Gaussian
representation
of
the
distribution
you've
seen
over
the
last
year
and
then,
given
that
you
can
then
answer
the
question.
What's
the
likelihood
of
seen
a
move
of
more
than
one
percent
in
one
day,
you
have
a
look
at
the
the
the
left
side
of
the
distribution,
which
is
the
positive
side
and
I'll.
B
Tell
you
what
an
up
movement
you
might
probably
see
only
three
percent
of
the
other
time,
but
in
terms
of
a
down
movement,
you
might
find
it
more
like
five
percent
of
the
time.
So,
given
the
percentage
movements
over
a
specific
day,
you
can
go
find
what
these
probabilities
are.
Now.
What
we
want
to
do
is
reverse
this
around.
We
want
to
start
with
the
probability
and
go
find
out
what
the
percentage
is.
B
So
we
want
to
say
you
know,
given
this
example:
I've
just
mentioned,
with
respect
to
be
the
normal
Gaussian
distribution
bell
curve
distribution.
You
can
ask
the
question:
what
is
the
chance?
What
is
the
extent
the
extent
the
loss
or
the
extent
gain
that
we
will
only
see,
maybe
one
every
twenty
days
or
one
every
hundred
days?
In
other
words,
what
is
the
level
that
we
can
sort
of
peg
in
the
sand
where
anything
that
happens
below
that
has
a
1%
chance
of
happening
or
a
5%
chance
of
happening?
B
Now,
if
you
have
a
look
at
the
distribution,
you
go
to
the
tails
and
you
add
up
95%
from
left
to
right
and
whatever
value
come
up
with
at
the
end,
that'll
be
the
number
that
you
see
now
if
we
used
our
simple
example
of
a
mean
of
let's
say
north
percents
and
your
your
volatility
of
1%,
then
your
ninety-five
would
be
around
about
sort
of
your
1
points.
6
percent
range
to
1.96
percent
range.
I.
B
Think
two
point:
three
three
is
one
percent
that
slips
my
mind,
sorry,
but
you
would
say
that
the
probability
of
seeing
something
happening
greater
than
one
point.
Six
percent
in
one
day
is
a
five
percent
probability
or
something
that
goes
further
than
two
point.
Three
three
percent
in
one
day
has
got
a
1%
probability.
Now,
once
you
figured
what
that
is,
then
that
actually
starts
becoming
a
little
bit
more
interesting
for
us,
because
now
we
know
what
the
extent
of
the
distribution
we
have
to
cater
for,
because
the
liquidation
ratio
says
well.
B
If
something
goes
wrong,
how
much
of
a
buffer
do
we
need
to
ensure
that
we
can
get
out
of
the
position
and
still
realize
the
full
value
of
the
loan
that
was
made
from
the
CDP?
So
if
you
have
a
look
at
eath,
for
example,
a
simple
calculation
would
be
something
called
Value
at
Risk
and
Value
at
Risk,
the
the
driving
params
underneath
that
is
actually
volatility.
You
do
this
thing
that
I
just
mentioned
to
you
create
your
future
curve.
B
You
find
the
extent
that
ways
up
what
the
the
percentage
of
risk
is
at
a
1%
level,
and
that
might
tell
you
that
over
one
day,
you've
got
a
chance
of
losing
two
point:
three
three
percent.
Now,
if
the
liquidation
ratio
needs
to
Cates
adjust
for
that,
then
that
equitation
ratio
would
be
102
point
three,
three
percent.
Why
is
that?
Because
when
the
CDP
gets
Brittany
and
you
purchase
that
collateral
and
you
want
to
basically
sell
it,
you
are
surmising.
B
That's
going
to
take
you
just
one
day
to
get
rid
of
all
that
collateral,
and
let's
pretend
for
a
moment
that
liquidity
is
there
for
you
to
be
able
to
do
so.
So
the
only
thing
that
you're
exposed
to
is
the
fact
that
there's
a
1%
chance
that
you
could
lose
two
point:
three
three
percents
of
your
of
the
value
of
your
collateral
at
the
time,
but
because
you
have
two
point
three
three
percent
more
well,
it
should
be
more
like
two
point:
four
five
percent
of
compounding
you
will.
B
You
should
have
enough
buffer
to
get
rid
of
your
collateral
and
realize
your
four
hundred
percent
of
your
of
your
loan,
and
that
is
wrapped
up
in
the
probability
of
1
percent.
So
you've
got
a
one
percent
chance,
theoretically,
speaking
from
the
calculations
that
we've
done
of
losing
more
than
two
point:
three
three
percent,
so
that
is
the
volatility
component
that
goes
into
liquidation
ratio.
It
gets
adjusted
by
the
liquid
the
the
liquidity
risk
component
because
you
might
say
well.
B
So
there's
this
this
weighting
that
you
have
to
consider
the
one
is
that
if
you
just
drop
everything
right
now
sure,
but
you
expose
yourself
to
the
the
inherent
liquidity
risk
and
if
it's
a
very
big
CD,
you
might
even
have
a
little
bit
of
endogenous
impact
as
well.
In
other
words,
you
may
force
more
liquidity
risk
which
will
go
against
the
the
just
the
volatility
that
you've
calculated,
so
the
liquidity
risk
might
add
on
an
extra
50
basis
points
or
maybe
1%.
So
you
start
off
with
two
point.
Three
three
add
on
one.
B
That
means
you've
got
33.33%
that
you
need
to
add
on
to
the
hundred
percents,
learn
that
you
take
a
naught
and
that
becomes
your
liquidation
ratio.
But
then
you
say
to
yourself:
well,
you
know:
you've
only
considered
one
day,
you've
only
considered
the
fact
that
you're
trying
to
get
rid
of
this
collateral
over
one
day,
that's
kind
of
silly,
because
everybody
needs
to
have
a
considered
way
of
trying
to
get
rid
of
the
collateral.
B
You
need
to
expand
that
time
period
a
little
bit
more,
so
that
folks
can
have
a
more
concerted
effort
in
getting
rid
of
their
collateral
and
at
the
same
time
you
need
to
make
sure
that
you
can
look
out
to
the
system.
So,
instead
of
a
be
headed
over
one
day,
you
may
look
over
to
look
at
it
over
seven
days
and
that's
pretty
much.
The
the
number
I've
picked
and
I
can
go
into
that
number
some
other
day
because
there's
a
bit
of
a
an
argument
that
one
can
form
from
that.
B
But
looking
at
seven
days,
you
could
then
increase
the
the
the
buffer
that
you
require
by
using
Gaussian
the
square
root
of
seven.
So
that'll
take
you
to
around
about
nine,
maybe
ten
percent,
so
your
liquidation
ratio
for
this
little
example
might
be
a
hundred
and
ten
percent
now
I'm
sure
right
now,
people
have
a
million
questions
or
may
feel
a
little
bit
baffled,
so
I'm
happy
to
go
moving
it.
You.
A
Know
I
have
a
few
questions
and
I
thought
where
I
know
you
have
a
heart:
stop
when
we
land
four
minutes
left
so
I,
don't
know
if
we
have
time
to
dig
into
this,
but
the
way
that
we
we
deal
with
C's
collateral
and
we
attempt
to
liquidate
it
in
the
CDP
mechanism
is
right
now,
through
a
direct
sale
on
the
dashboard
and
in
the
future
it
will
be
through
a
rather
sophisticated
auction
mechanism.
So
when
you're
predicting
the
liquidity
of
a
particular
asset
class
you're,
also
the
risk
teams
also
have
to
factor
in
that.
A
B
I
mean
the
first
aspect
about
this
is
saying
that
you
excuse
me
my
computer,
just
weird
on
me.
The
these
two
aspects,
the
one
is
in
the
liquidation,
auction
you
buy
that
collateral
right
and
hopefully
you
get
some
sort
of
benefit
from
it,
ie
a
discount
and
then
once
you
have
the
collaterally
in
hand.
What
are
you
going
to
do
with
it
now
in
the
crypto
space?
This
is
becomes
a
lot
more
interesting
because
in
the
traditional
space,
what
you
do
is
someone
defaults
on
their
house.
B
B
Have
to
go
from
that
collateral
into
USDollar
interfere.
You
could
go
from
that
collateral
into
maybe
the
native
token
of
the
blockchain,
because
it
should,
theoretically
speaking,
be
more
liquid
and
be
the
representative
currency
for
that
particular
system.
The
other
thing
you
can
consider
is
all
other
collateral
or
all
other
tokens
that
may
have
more
liquidity
than
it.
B
So
you
might
come.
You
might
see
that
a
CDP
gets
bitten
and
you
purchase
token
ABC.
But
do
you
find
that
there
is
about
seven
tokens
that
are
much
more
liquid
in
it
and
you
may
find
that
there
are
many
avenues
that
you
could
have
a
look
at
in
terms
of
trying
to
get
rid
of
that
particular
token
into
those
other
more
liquid
assets.
B
A
B
You
might
find,
as
people
want
to
you
know,
get
rid
of
their
collateral
need
to
die.
Wow
yeah,
it's
complex,
so
it's
very
interesting,
but
the
complexity
gives
you
opportunity.
An
opportunity
means
that
you've
got
all
these
very
intricate
possibilities
of
arbitrage
in
you've
got
all
these
intricacies,
where
you
can
see
possibilities
of
providing
liquidity
to
systems.
It's
just.
The
whole
thing
is
interacting
to
such
an
extent
that
if
it
is
done
appropriately,
there's
this
sort
of
net
benefit
this
virtuous
circle
of
benefit
that
you
actually
employ
into
the
whole
system.
B
The
crypto
space
is
so
interesting
because
of
all
the
different
vectors
you
can
take.
The
the
different
granularities
in
terms
of
interpreting
tokens
will
give
you
another
sense
of
it.
What
happens
if
you
are
getting
a
utility
token
you?
You
might
want
to
get
a
utility
token
and
sell
it
out
for
a
security
back
token,
because
you've
got
a
different
play
between
utility
and
stocks.
You
might
have
a
different
play
between
a
governance,
token
and
a
currency
target.
There's
so
many
ways.
B
A
Well,
that's
another
mind-blowing
concept
to
close
the
meeting
on
I
think
that
there's
David
and
Jordan
you
had
some
interesting
questions
on
we're.
Not
gonna
have
time
to
get
to
them
today,
but
I
got
to
put
it
our
dog
for
the
next
meeting
and
we'll
get
to
them.
Then
thanks.
That
was
amazing.
Thank.