►
From YouTube: Governance and Risk Meeting: Ep. 17
Description
In this Governance and Risk meeting with Steven Becker we continue our discussion on Risk Teams and ways to incentivise them.
Meeting Minutes: https://github.com/scottrepreneur/mak...
Podcast version: https://soundcloud.com/makerdao/sets/...
Governance and Risk Framework - Part 1: https://medium.com/makerdao/makerdao-...
Governance and Risk Framework - Part 2: https://medium.com/makerdao/makerdao-...
Foundation Proposal: https://medium.com/makerdao/foundatio...
Governance and Risk Framework - Part 3: https://medium.com/makerdao/makerdao-...
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
Hello,
everyone
and
welcome
to
the
first
scientific
governance
and
risk
meeting
of
2019.
So
this
is
a
very
exciting
time
for
me,
because
I've
missed
these
meetings,
a
lot
I
feel
like
I'm,
losing
track
of
what's
happening
and
make
her
doubt
if
we're
not
having
a
scientific
government
and
bris
meeting
at
least
once
a
week,
so
half
will
hear
about
it,
got
it
Stephen
I
gonna
hand
it
off
to
you
and
let
you
that
wasn't
on
the
agenda.
No.
B
B
That
being
said,
I
also
refer
to
these
risk
teens
as
those
that
support
and
those
that
submit
now.
That
was
a
little
bit
of
a
misnomer
with
respect
to
the
support,
because
effectively
both
risk
teams
did
submit,
and
this
brings
us
him
to
you
know
the
conversation
for
today.
You
know
what
are
supporting
teens,
you
know,
who
are
they
and
even
more
importantly,
is
if
they're
supporting
how
should
they
be
compensated.
So,
let's
start
at
the
beginning,
so
what
do
I
mean
by
supporting
risk
teams?
B
A
B
B
Now
keep
those
to
keep
that
difference
in
mind
and
keep
those
two
aspects
in
mind
now,
given
that
you
use
it
as
collateral,
and
given
that
a
CDP
is
liquidated
and
you
try
to
liquidate
the
online
collateral,
you
can
either
obviously
go
into
a
different
sort
of
liquidation,
vector
like
you
could
use
the
collateral
go
purchase
eath.
We
can
use
collateral
and
go
in
and
and
purchase
dye,
whatever
the
case
may
be.
But
if
everything
fails
it
boils
down
to
well,
how
do
you
go
back
to
the
issuer
and
claim
against
them?
B
Now
we
need
to
be
able
to
you
boil
that
down
to
the
granular
level
of
someone
has
done
an
analysis,
and
someone
has
found
that
this
is
the
part
that
you
should
take
so
very
much
like
Richard
says
about
digits.
If
you
do
have
digits
collateral
and
you
want
ultimate
recourse,
it
really
does
until
you
travel
into
Singapore
presenting
yourself,
then
you
get
this
little
gold
bullion,
which
is
this
tiny
little
piece
of
gold,
which
is
actually
pretty
cool.
B
You
should
learn
to
this
size
and
have
a
look
at
it,
and
but
that
doesn't
end
it
there,
because
all
that's
done
is
giving
you
gold
into
your
possession
recourse
all.
That
would
then
illuminate
the
reality
of
that.
Just
because
you've
got
gold,
in
your
hand,
doesn't
mean
you've
actually
have
full
liquidation,
because
you
still
need
to
turn
it
into.
Unfortunately,
Singaporean
dollars,
not
just
US
dollars
or
euro
or
pounds.
Now.
How
do
you
do
that?
Because,
as
soon
as
you
get
that
gold
into
your
hand,
you've
taken
it
out
of
the
integrity
process?
B
Now
you
have
to
get
back
into
the
integrity
process,
which
means
that
you
either
got
to
submit
that
gold
back
in.
If
you
wish
to
use
it
again
within
the
the
token
construct,
or
you
simply
have
to
go
to
a
designated
broker,
which
is
very
important
and
be
able
to
sell
that
gold
directly
to
them
now,
I,
don't
know
if
any
of
you
guys
have
tried
this
we're
taking
a
piece
of
gold
in
your
hand
and
walking
to
your
brokers,
office
will
say
and
I
want
to
sell
this
to
you
not
that
easy.
B
You
really
have
to
make
sure
that
your
paperwork's
in
place
and
everything
is
very
well
documented
before
you
can
present
that
that
bullion
de
liquidating
in
Sofia,
so
these
recourse
orders
are
going
to
require
legal
teams
they're
going
to
require
folks
that
really
look
into
this,
especially
when
we
consider
non
fungible
tokens.
What
about
invoices?
What
about
other
possibilities
that
could
be
turned
into
into
non
four-hundred
serpents
or
if
you
have
a
collection
of
non
fungible
tokens
that
have
been
made
fungible
and
you
use
it
as
collateral
as
well.
B
B
The
body's,
probably
more
than
aware
of
on
this
call,
it
really
boils
down
to
a
couple
of
things.
One
we
are
looking
for
a
RC
20
compliant
circuit.
Is
it
in
fact
the
case?
Then
you
go
further
to
have
a
look
at
any
sort
of
operational
risks
that
lie
within
the
coating
of
the
of
the
in
token
itself,
but
I'm
not
gonna,
bore
you
guys
with
that.
B
I
think
you
guys
probably
know
that
pretty
better
than
I
do,
but
in
effect,
you're
supporting
teams
have
to
cover
this
wide
spectrum
of
coloured
verticals
or
risk
verticals,
and
what
we
got
to
figure
out
is
how
do
we
vet
these
teams?
How
do
they
contribute
and
how
do
those
control
their
contribution
get
compensated
because
there's
this
idea
that
you
know
I
mentioned
beforehand,
where
you
get
a
collection
of
risk
teams
working
together,
it
may
work,
but
ultimately
you
might
find
it
both.
B
You
might
find
that
compensation
really
is
in
the
hands
of
the
person
or
the
team
that
submits
the
risk
parameters,
and
then
that's
about
it.
Look
it's
all
about
negotiation
in
the
background
that
might
work
or
that
might
break
the
system.
I'm,
not
too
sure,
there's
another
Avenue,
where,
if
you're
a
supporting
risk
team,
you
actually
submit
what
you
are.
B
You
submit
the
data
you
submit
the
work
that
you
have
that
actually
supports
all
the
other
versus
teams
that
wish
to
actually
generate
risk
parameters
for
the
system.
So
how
do
we
go
about
doing
that?
As
an
example,
you
have
a
a
support
team
that
gathers
data,
cleans
it
and
provides
it
to
to
the
system
how
they
compensated.
B
Do
we
actually
should
we
require
them
to
have
the
data
made
available
to
everyone
open
source
or
do
they
keep
it
for
themselves
and
then
interact
with
the
risk
team
that
actually
submits
the
the
parameters
at
the
end?
If
so,
what
are
the
economics
in
there?
How
does
it
work
and
what's
the
the
dynamics
between
the
teams
that
try
and
provide
the
risk
parameters
at
the
end
of
the
day,
like
I,
feel,
like
I,
said
a
whole
lot?
There
I'll
stop
for
a
moment
to
see
if
there's
any
questions
that.
A
Might
be
a
good
opportunity
for
me
to
remind
everybody
I
forget
to
do
this
at
the
beginning.
The
call
if
anybody
has
any
questions,
please
feel
free
to
jump
in
at
any
time.
We
actively
encourage
it,
don't
consider
it
into
an
interruption
if
you
don't
have
access
to
a
microphone
type.
Your
questions
in
the
chat
at
the
side
of
the
window
and
I'll
read
them
out
when
there's
an
opportunity.
I'm
Stephen
one
of
the
questions
I've
had
and
stop
me
if
I'm
jumping
ahead,
but
well,
I'm,
always
kind
of
curious
or
never
or
troubling.
A
Talk
wrap.
My
head
around
word
of
risk
teams
come
from
it,
or
is
this
a
concept
that
already
exists
in
the
financial
world
that
you
know
you
can
go
to
company
X
and
they
have
a
risk
team
internally?
Is
that
what
they
call
it?
Or
is
this
something
that
we're
kind
of
inventing
in
the
crypto
space?
Oh
no.
B
There's
nothing
new
about
this.
Everything
to
do
with
risk
is
really
old
hat.
It
really
is
a
case
of
taking
the
traditional
tools
and
applying
it
to
the
space,
but
obviously
we
need
to
tweak
it
appropriately
and
we
need
to
add
in
a
few
dynamics
that
are,
you
know
clearly
not
present
in
the
traditional
space
number
one,
the
most
obvious
one
is
a
decentralized
risk
function
that
is
I
would
say.
To
my
mind,
unheard
of
that's
the
one
thing
that
we,
you
know
that
we
truly
innovating
on,
so
that
requires
consensus.
B
That
requires
the
ability
to
figure
out
how
to
make
this
already
particularly
difficult
function
in
the
traditional
world
function
in
our
world
by,
firstly,
splitting
it
up
into
component
parts
that
everyone
understands
and
then
sort
of
abstracting
it
into
a
general
application
that
works
in
the
traditional
world.
If
there's
something
called
Enterprise,
Risk
Management,
which
literally
says
take
care
an
organization,
look
at
every
possible
risk
thing
that
could
happen
to
the
risk
event
that
could
happen
within
it
and
try
and
manage
it.
B
It
is
a
massive
subject
matter
and
it
really
entails
going
down
to
the
details
of
processing
transactions.
What
does
it
look
like
how
to
audit
it
how
to
very
file
validate
it
and
so
on
and
so
forth?
So
the
tools
are
there.
It
really
is
a
case
of
bringing
it
over
to
to
our
space,
using
the
tools
that
work
now
and
then
eventually
seeing
how
we
can
expand
this
to
bring
in
the
rest
of
the
tools,
obviously,
with
the
the
innovation
for
for
our
space
included.
I,
don't.
B
That
there
are
from
the
the
big
four
auditors
that
have
their
own
departments
for
risk
management
all
the
way
through
to
independent
companies
all
the
way
through
to
large
financial
institutions
and
even
investment
banks
that
outsource
their
risk
management
function.
It
is
absolutely
everywhere,
and
you
find
that
you
know
the
processes
that
are
defined
within
these
big
institutions
are
extraordinarily
detailed
and
they
are,
you
know,
well
documented.
It
would
be
great
to
see
someone
in
that
space.
B
Take
that
massive
process
and
try
and
bring
it
over
to
to
outside
and
actually
try
and
implement
it,
and
that
is
effectively
what
I'm
trying
to
do.
But
there
is
so
much
of
it
that
if
I
had
to
basically
just
drop
it
all
right
now,
it
really
would
be
like
a
ton
of
bricks
on
top
of
our
heads
and
most
of
it
wouldn't
really
be
applicable
because
we're
not
at
that
breadth
and
depth
of
the
economy
or
of
market
to
be
able
to
use
all
these
tools.
The
granularity
is
absolutely
you
know.
B
Massive
to
a
certain
extent
is
actually
quite
interesting
and
ironic.
The
granularity
has
been
taken
care
of
because
of
the
immutability
of
the
blockchain,
and
it
really
sort
of
boils
down
to
grabbing
that
data
and
repurposing
it,
for
you
know,
risk
man,
a
risk
management
function
and
in
doing
that,
you
can
really
enlighten
yourself
as
to
how
your
organization
is
functionally
not
only
with
in
itself
with,
but
with
like
within
the
actual
blockchain
space
as
well.
We
are
in
a
much
better
position
to
really
suck
out
more
value
from
these
traditional
tools.
B
Then
you
know
folks
in
the
traditional
space.
I
cannot
give
use
the
word
traditional
apart
too
many
times
now.
That's
the
so
that
effectively
is
a
rundown
of
the
supporting
teams.
Cuz
I
mention:
what
do
we
do
with
respect
to
the
data
they
create
and
how
do
we
basically
compensate
them
appropriately?
Now,
I,
don't
have
the
answers
here.
I
mean,
are
good
ideas
but
I.
B
So
if
you
don't
have
the
answers-
and
this
is
something
that
I'm
sort
of
putting
out
there-
and
you
know
focusing-
is
chuh
and
think
about
this-
and
you
know
come
back
to
us,
but
if
you
are
a
risk
team
that
sources
data
cleans
the
data
and
makes
it
available.
The
one
thing
that
the
the
nkr
token
holders
or
governance
needs
to
do
is
vest
that
process.
Where
are
you
getting
the
data
from
how
you
clean
in
it
and
how
are
you
providing
it
now?
What
do
I
mean
by
a
provision
of
data?
B
Is
this
data,
and
this
is
the
detail
that
we
need
to
really
think
about.
Is
this
data
centralized
to
this
one
risk
team,
or
is
this
data
made
available
in
a
in
a
decentralized
a
platform
like
ipfs
or
what's
other
one
storage,
or
something
like
that,
where
everybody
can
have
some
sort
of
accessibility
to
it?
I'm
not
sure
now,
I've,
given
that
you'll
find
that
submitting
risk
teams
would
say:
well,
that's
fantastic
I
would
use
it
and
then
they
would
use
the
information
create,
see
the
the
models
for
them.
B
So
the
crazy
models
for
themselves
use
the
data
and
obviously
come
up
with
risk
parameters
that
they
would
contribute
towards
the
system.
Now
their
submission
is
very
clear
and
therefore
the
compensation
will
be
very
care,
but
what
what
do
we
say
about
the
guy
that
submits
or
the
riskiness
submits
the
data?
What
do
we
say
about
Contribution?
What
we
say
about
how
they
get
compensated?
Is
there
a
fixed
rate
process?
I.
Think
rich.
You
mentioned
this
in
the
previous
meeting.
Is
there
a
fixed
rate
process?
Is
there
a
there
a
compensation
per
submission?
B
You
know
what
is
this?
What
is
this
economic?
That's
going
to
make
this
work
appropriately.
Now
the
as
I
said:
I
have
ideas,
but
I
want
to
put
that
out
there,
for
you
know
folks
to
contribute
and
and
sort
of
give
some
suggestions,
but
that
is
if
that
is
one
aspect:
the
data
infrastructure.
What
does
it
look
like
not
just
with
respect
to
quantitative
data,
but
with
respect
to
qualitative
data?
One
aspect
that
we
definitely
need
is
qualitative
assessment,
so
you
might
have
risk
teams
that
do
this
qualitative
assessment.
B
B
And
what
I'm
going
to
to
ask
is
that
if
anyone
does
have
any
ideas
feel
free
to
to
do
to
drop
an
idea
feel
free
to
to
email
in
you
know
anything
is
welcome
now.
The
other
aspect,
which
is
also
important,
is
you:
have
all
these
risk
teams
and
they're
all
contributing.
Now
we,
let's
pretend
we've
defined
very
well,
how
they
contribute
and
well
this
data
is
stored
and
how
this
data
is
accessible
and
you
can
use
it.
How
do
these
RIF
teams
communicate
between?
B
How
do
you
communicate
in
a
way
where
you
have
open
dialogue
that
is
transparent
to
the
community?
To
understand
the
cool
of
the
mind
of
the
risk
function?
You
we
need
to
centralize
the
communication
we
need
to
provide
a
platform.
Is
that
platform
going
to
be
next
to
the
the
waiting's
that
are
submitted
to
the
risk
teams,
where
we
aggregate
the
consensus
on
the
risk
parameters
again,
this
is,
you
know,
obviously
a
very
important
aspect
to
consider
so
again,
I'm
going
to
pause
there
for
a
second
to
see,
if
there's
any
digestion
that
is
required.
A
So
what
kind
of
mechanisms
are
we
gonna
put
in
place
to
start
gathering
that
kind
of
feedback
or
or
I
guess
kind
of
really
solidifying
what
the
risk
teams
would
be
like,
and
if
people
were
interested
in
becoming
a
part
of
a
risk
team
who
would
they
go
to
or
what
are
the
barriers
to
entry
things
like
that?
Like
do?
We
have
sort
of
a
any
thoughts
about
what
a
process
to
get
this
thing
kicked
off,
would
look
like,
or
do
you
have
something
kicked
off
in?
A
B
That's
that's
what
I'm
doing
is.
You
know
purposefully
eliciting
a
lot
of
thinking
with
respect
to
how
does
one
become
a
risk
team?
Now?
What
are
the
things
to
consider,
because
the
more
what
I'm
hoping
is
that
the
more
I
talk
about
this
particular
topic?
You
know
and
and
split
it
down
into
its
granular
form.
You
might
find
there's
one
aspect
that
resonates
with
someone
and
that
person
or
that
team
will
go
and
the
second
we
could
quite
easily
do
that.
B
In
fact,
we
might
be
able
to
create
a
business
of
the
back
of
that
now,
going
about
forming
a
risk
team
worrying
about
contributing
we,
the
internal
risk
team
of
the
foundation,
purposely
going
to
be
the
model
and
we're
not
gonna,
be
the
muddle,
as
already
in
terms
of
the
benchmark
of
the
best
kind
of
risk
management
in
the
world,
because
that's
impossible.
That
requires
a
hell
of
a
lot
more
resources
and
we
have,
but
what
it
does
do.
B
It
gives
a
very
well-defined
path
on
how
you
become
a
risk
team,
how
you
get
bested
by
a
community,
how
you
getting
included
into
the
system
and
how
you
go
about
contributing
towards
not
only
the
data
but
the
data
models,
model
validation
and
you
know
risk
parameters
as
well.
I
mean
the
whole
thing
is
very
much
inclusive,
so
the
sneaky
part
about
this
is
that
now
I'm
trying
to
put
out
there
the
idea
in
everyone's
minds
that
anyone
can
be
a
risk
person
or
risk
team.
B
It
really
boils
down
to
what
you
want
to
contribute
and
really
what
that
is
going
to
look
like
is
well.
Follow
the
lead
internal
foundation,
risk
team
and
see
how
they're
forming
and
see
how
they
are
are
sort
of
contributing
towards
the
system.
Now,
there's
the
one
aspect
that
I
I
didn't
speak
about,
or
rather
one
group
of
regimes
I
didn't
speak
about-
are
be
to
submit
it.
B
The
guys
that
actually
put
forward
models
now
we've
been
harping
on
for
the
longest
time
about
collateral
models,
now
put
forward
a
model,
so
you
can
calculate
the
risk
parameters
for
this
particular
circuit,
very
simple.
But
what
we
are
missing
are
two
more
groups
of
models,
and
this
I
haven't
touched
on
yet
and
for
the
new
year.
I
think
this
might
be
another
new
subject
matter
that
is
open
for
discussion.
The
one
is
portfolio
models
and
we've
been
speaking
about
collateral
models
and
understanding
them,
but
what
about
the
portfolio
as
a
whole?
B
We
obviously
need
folks
to
to
look
at
the
model
and
say
well
the
collateral
portfolio
we
have
has
contributed
so
much
stability
to
the
system.
How
is
that
quantified?
Where
is
that
metric?
Look
like
what
is
the
diversification
benefit
that
we
have
achieved?
Can
we
actually
create
more
with
the
business
needs
of
the
system
as
well?
The
the
other
set
of
models
are
the
sort
of
economic
models
I
mean,
as
you
probably
witnessed,
and
this
might
be
actually
a
segue
into
another
discussion.
B
We've
recently
decreased
the
stability
fee.
Now,
there's
a
lot
to
be
said
about
why
that
happened.
What's
the
rationale?
What's
the?
What
does
it
look
like
going
forward?
How
do
we
actually
rationalize
these
increases
and
decreases?
And
what
does
it
look
like
in
terms
of
the
model?
Well,
that's
one
of
the
aspect
that
a
risk
team
contribute
is
an
economic
model
and
it
might
be
on
a
case
of
saying
well,
we've
studied
the
we've
studied
this
space,
and
we've
noticed
that
there
is
a
massive
blood
of
inventory.
B
That
means
we
need
to
increase
the
the
list
ability
fee
there
is,
you
know,
a
massive
demand
for
it,
but
not
enough
supply.
What
we
need
to
decrease
the
either
stability
peep.
What's
the
model
behind
that
where's,
the
data
coming
from
the
processes
are
very
easy,
but
it's
the
data
and
the
empirical
justification
for
it.
That's
a
little
bit
more
difficult.
So
there
are
two
other
aspects
that
I
have
never
touched
on
before
the
one
being
is
the
portfolio
models
and
economic
models
and,
together
with
the
either
submitting
team,
specifically
recourse
in
technical
with
it.
B
B
B
If
folks
think
that
this
is
slightly
confusing,
I
feel
so
happy
to
elaborate
on
any
point,
but
that
fully
intents
and
purposes
is
what
we
are
going
to
look
at
and
what
we're
going
to
deal
with
this
year,
you're
going
to
find
that
the
the
risk
function
is
going
to
come
front
and
center
a
lot
more
than
it
has
in
the
previous
year,
I
think
we
were
setting
the
stage
and
the
foundations
for
it.
Now
we're
going
to
start
implementing
it.
Now
we're
going
to
start
answering
questions
like
how
do
I
become
a
risk
team.
B
How
does
one
submit
risk
parameters?
What
does
the
data
look
like?
How
do
I
submit
the
data
and
then,
at
the
same
time
not
only
asking
questions,
but
maybe
in
providing
answers
such
as
I
would
like
to
submit
data
to
specialized
storage
function.
I
would
like
to
create
a
qualitative
assessment
model
that
includes
certain
you
know
traditional
aspects
and
so
on
so
forth.
B
A
B
A
I
was
talking
about
how
you
kind
of
throw
things
up
to
the
community
for
for
answers,
and
it
I
guess
this
is.
The
sense
is
when,
should
we
start
thinking
about
aggregating
that
data
in
earnest
so
is?
Is
there
a
rough
road
map
of
what
the
risk
teams
are
going
to
look
like
and
how
they're
going
to
evolve
over
the
course
of
this
year,
or
it
is
the
first
step
to
start
reaching
out
to
the
queue
getting
that
kind
of
event?
Oh
no.
B
B
Suppose
it's
coming
to
fruition
in
a
sense
that
all
the
stuff
that's
been
written
discussed
is
going
to
become
more
open
in
terms
of
how
to
actually
go
about
doing
calculations.
What
the
underlying
idea
is
and
so
on
and
so
forth.
That's
going
to
be
the
first
thing
that
comes
on
the
third,
the
third
article
that
was
submitted
on
on
risk
governance,
looked
at
the
broad
application
of
governance,
but
I
did
mention
at
the
the
bottom
of
that
of
that
article
that
there's
going
to
be
I,
wouldn't
say
technical
submission.
B
It's
just
it's
going
to
be
a
submission
with
the
math
in
it.
It
explains
how
we
go
about
looking
at
a
portfolio
of
collateral
and
understanding
the
application
of
something
called
liquidity,
value
and
liquidity
adjusted
body
at
risk.
That's
it!
So
it's
going
to
be
simply
a
look
at
that,
but
folks
can
read
that
in
a
process
at
a
bit
and
probably
come
up
with
better
ideas
or
effectively
just
try
and
implement
it
themselves,
and
you
know
Python,
R
or
whatever
the
case
may
be.
But
this
is
a
two-track
process.
B
I'm,
certainly
not
waiting
for
the
communities
come
back
with
ideas
first
and
then
respond
it's
going
to
be
definitely
parallel
because
I'd
like
to
get
the
dialogue
going,
I
want
to
get
the
conversation
going
and
in
order
to
have
a
conversation
we
know
if
you
first
need
to
have
no
very
strong
opinions
about
how
we
should
move
this
forward.
So
internally
we
have
a
very
firm
foundation
of
what
we'd
like
to
do,
but
now
it's
going
to
be
a
case
of
saying.
Well,
you
know
we've
spoken
about
this
for
a
while.
B
Now
now
it's
almost
like
the
community
is
trying
to
say
okay.
Well,
this
is
what
I
would
like
to
do
and
then
we
start
bashing
ideas
together,
and
so
we
actually,
almost
through
dialogue
and
debate,
make
clear
to
the
community
what
the
process
looks
like,
because
I
find
that
if
you
just
list
the
process-
and
you
say
you
know
effectively-
it's
a
simple
step-
one
to
step-
10
and
that's
how
you
do
it.
B
It
seems
like
a
good
idea
until
you
try
doing
it,
and
you
realize
there's
a
lot
of
nuances
and
to
try
and
figure
out
all
the
permutations
of
the
nuances.
Is
it's
almost
impossible:
you'll
stall,
the
whole
procedure,
but
if
you
do
it
together
and
you
actually
discover
the
nuances
together,
the
ones
that
we
already
anticipate
and
even
new
ones
that
we
didn't
anticipate.
B
You
can
move
through
these
steps
and
truly
get
to
your
final
conclusion
and
you've
done
that
organically
we've
done
that
organically
and
you've
involved
community
and
you've
got
a
better
sense
of
how
you're
actually
addressing
not
only
just
risk
but
the
stability
in
the
entire
system,
because
this
is
a
very
holistic
problem.
It
doesn't
just
boil
down
to
what's
the
risk
of
this
token
I.
Stick
it
into
a
collateral
portfolio
and
that's
where
it
ends.
It
talks
to
everything
it
talks
to
you,
the
dislocation
of
supply
and
demand.
B
A
A
A
B
God:
okay,
no
understandable,
understandable,
okay!
So
then,
if,
if
there's
nothing
further,
then
I
can
just
you
know,
have
a
look
over
my
topics
that
I
wanted
to
cover.
It
was
214
teams
just
to
revise
the
maps.
You
know
guys
that
provide
quantitative
and
qualitative
data
data.
Sorry-
and
that's
also
when
I
say
qualitative
data
I'm
talking
about
all
aspects
of
data
that
allows
for
for
risk
assessment,
you
can
have
wrist
support
team
that
looks
at
not
only
data
cleaning
but
validation
services,
model
validation,
services
that
actually
put
an
important
aspect
as
well.
B
You
know
it's
all
very
well
for
folks
to
submit
models
for
NPR
governance
to
consider,
but
you
might
find
that
there's
very
few
people
that
can
actually
you
know,
work
that
model
appropriately
to
its
final
point,
you
may
need
some
sort
of
validation
service
recourse
orders,
as
we
mentioned,
be
forged
and
they're
actually
very
important,
especially
coming
into
the
the
new
year
with
entities
that
are
going
to
come
on
board.
You're
gonna
have
security
tokens
gonna
come
on
board
the
whole.
B
The
tokenization
of
security
is
the
whole
stack.
We
have
to
get
our
head
around
to
understand
how
that
is
actually
going
to
not
only
apply
not
only
apply
with
respect
to
that
particular
industry
forming,
but
also
with
respect
to
how
collateral
can
be
used
within
the
the
nacre
system.
I
technical
that
you
guys
know
better
than
I
do
so.
I'm
not
gonna,
go
into
too
much
detail
there.
B
As
far
as
the
the
risk
submitting
sort
of
the
teams
that
are
submitted,
we've
always
spoken
about
collateral
models,
but
I'd
like
to
start
stretching
everyone's
mind
to
portfolio
models
and
economic
models
as
well,
because
all
of
this
comes
together.
You'll
find
that
as
we
progress,
it's
going
to
be
pretty
obvious
that
certain
things
need
to
be
addressed.
You
start
off
with
collateral
collateral
type.
You
realize
you
need
to
look
at
the
portfolio.
You
start
at
the
portfolio.
You
realize
you
need
to
look
at
the
system.
You
look
at
this
system.
B
You
realize
you
need
to
understand
the
economics
of
it,
so
the
thing
really
relates,
even
if
you
think
that
all
you're
doing
is
calculating
a
simple
stability
fee
for
collateral
size
or
a
simple
debt
ceiling
for
a
collateral
type.
But
in
fact
what
you're
doing
is
implementing
a
parameter
and
or
feature
that
actually
affects
the
whole
system.
A
That
part
I,
find
this
special,
especially
fascinating,
is
the
portfolio
management.
Now,
what
are
some
of
the
things
that
you're
looking
at
your
risk
team,
look
at
if
they're
taking
a
holistic
view
of
all
the
different
collateral
types
and
somebody
proposes
of
the
collateral
type?
Is
it
just
correlation
or
is
there
some
deeper
relationships
that
you'd
consider
well.
B
This
is
where
the
innovation
comes
into
our
space,
specifically
from
the
traditional
space.
It
was,
you
know,
very
simple:
it's
a
risk
return,
optimization
calculation,
where
you've
got
expected,
returns,
expected
risks
and
notes.
I
use
the
word
expected,
which
means
is
the
statistical
element
to
it,
and
then
obviously
you
have
you
expect
correlation
matrices.
B
If
you
think
of
a
large
mutual
fund,
this
decides
that
need
to
allocate
five
and
a
half
billion
dollars
to
Apple.
You
know
it's
not
going
to
go
into
the
market
quickly
in
searches,
five-and-a-half
billion
dollars
of
apples,
and
it
is
very
slowly
because
it
doesn't
want
to
take
the
price.
This
is
where
the
liquidity
adjustment
comes
into
into
consideration
the
same
with
us.
So
we
have
a.
We
have
a
portfolio
of
collateral
and
we've
got
three
types
of
debt
ceilings.
B
The
broader
debt
ceiling,
the
theoretical
debt
ceiling,
tells
you
well
and
if
everything
was
good
in
the
whole
world,
it
was
great.
How
much
could
we
take
of
this
particular
collateral
type?
But
in
order
to
get
there,
we
better
be
practical.
We're
gonna,
we're
gonna
have
a
look
at
it.
We
have
to
consider
it
in
terms
of
its
market
cap,
so
sort
of
considered
in
terms
of
its
liquidity.
We
got
considered
in
terms
of
its
price
volatility
and
then
on
top
of
that-
and
this
is
where
it
really
becomes
as
innovative
from
our
particular
space.
B
We've
got
to
consider
it
in
terms
of
its
business
function,
its
economic
function
to
the
to
the
remedy
of
any
dislocation
between
supply
and
demand
or
the
further
provision
of
supply,
and,
if
you
believe
in
you
know,
sort
of
supply-side
economics
to
actually
spur
demand.
So
there
are
there's
so
many
aspects
about
the
traditional
space
that
you've
taken
into
our
own
space
and
we
have
to
sort
of
meld
them
together
so
that
we
actually
create
not
only
a
risk
management
mechanism
but
a
business
development
mechanism
as
well.
B
So
it's
correct:
it's
the
the
risk
mechanism,
the
business
development
mechanism
and
an
economic
mechanism
at
the
same
time.
So
that's
what
a
a
portfolio
construction
from
our
point
of
view
looks
like
as
an
example
if
we
decided
that
I
always
like
to
use
digits,
because
I
think
everyone
can
relate
to
the
fact
that
it's
gold,
if
you
have
a
look
at
this
theoretical
foundation,
it's
gold.
So
therefore
people
would
think
well,
there's
gold,
you
know
it's
it's
pretty
done
liquid
out
there.
You
can
do
anything!
B
You
want
with
it
and
there's
approximately
nine
trillion
dollars
of
supply.
You
know
swishing
around
the
world,
so
we
can
have
this
mess.
That's
true.
Theoretically,
speaking,
practically
speaking,
the
people
that
provide
this
organization
provides,
as
did
Jax,
has
stand
to
be
corrected
at
this
point,
but
they've
got
about
three
million
dollars
worth
of
of
the
token.
So
here's
where
you
have
a
look
at
the
practical
aspects
of
them
say
well,
three
million
is
the
market
cap
of
this
particular
talk,
even
though
it
does
represent
goals.
B
That
is
a
nine
trillion
dollar
a
supply
out
there.
We
can't
consider
it
as
performing
theoretical
basis.
We
have
to
look
at
it
from
a
practical
basis.
Even
when
we
look
at
the
practical
basis,
we
can't
say:
okay,
let's
make
the
Dead
City
in
three
million.
We
can't
exact
a
full
market
cap.
We
then
need
to
consider
okay
well,
what
portion
of
it
can
we
can
reconsider
and
that
comes
down
to
you
know
what
sort
of
liquidity
aspects
you
want
you
to
be
subjected
to
now.
Why
am
I
talking
about
this?
B
Because
we
have
referred
to
this
before
we're
looking
at
this,
because
we
need
to
balance
the
fact
that
there
is
a
difference
in
correlation
that
is
required.
But
there's
also
a
business
reason
as
well.
To
be
able
to
include
digits,
might
be
able
to
facilitate
a
boot
or
a
boot
strapping
process
to
get
digits
being
used
more.
So
you
have
the
ability
to
actually
catalyze
or
have
the
ability
to
catalyze
it,
and
you
might
find
the
digits
grows
from
three
million
to
10
million
within
the
month,
just
because
it's
an
American
system.
B
B
To
a
certain
extent,
we
will
always
have
that
contributing
factor
to
to
a
token,
because
we
provide
two
of
the
primary
functions
that
drive
all
economic
systems,
the
ability
to
do
pledge
collateral
and
the
ability
to
to
draw
value
that
pretty
much
drives
everything.
So,
if
you've
included
this
within
your
ecosystem
of
your
token,
you
actually
are
going
to
be
able
to
capture
the
liquidity
premium
and,
at
the
same
time,
you're
going
to
be
able
to
get
a
source
of
financing
for
yourself
that
was
never
before
available.
B
So
these
two
aspects
are
going
to
always
play
into
an
organization's
thinking,
and
to
that
end
it
really
boils
down
to
how
big
that
organization
is
I
mean
if
we
have
a
very
large
organization
that
we
use
as
collateral
like
right
now,
it's
cerium
to
a
certain
degree.
We
do
contribute,
but
I
don't
say
we
are
the
primary
factor
towards
the
the
etherium.
A
success
story,
but
I
reckon
give
us
a
little
bit
more
time
and
we'll
we'll
be
one
of
the
primary
factors
to
this
am
success
story.
B
You're
gonna
find
that
you
know,
for
me:
cerium
is
every
time
I
read
more
about
the
details
of
it,
and
you
know.
I
must
confess
that
my
technical
aspect
is
not
terribly
strong
there.
It
invigorates
me
to
the
point
of
realizing
that
this
is
a
unique
network
that
has
the
ability
to
facilitate
a
completely
different
economy
that
will
just
dovetail
appropriately
with
the
traditional
one,
and
we
are
the
effector
primary
functionality
to
help
drive
a
financial
ecosystem
on
top
of
this
as
well.
B
A
I'm
just
trying
to
figure
out
how
about
what
a
corollary
and
the
traditional
market
there,
the
traditional
financial
world
would
look
like
if
because
make
it
out
kind
of,
we
occupy
a
kind
of
a
lonely
spot
in
the
ecosystem.
Right
now,
there's
there's
us
and
there
I
think
compound
sort
of
does
something
kind
of
like
what
we're
doing.
But
if
a
token
XYZ
comes
in
and
their
value,
we
accept
as
a
collateral
type
man
because
of
the
our
acceptance
their
value
doubles
overnight,
and
then
it
comes
time
for
something
bad
happens
in
the
market.
A
We
need
to
liquidate
a
bunch
of
CDP's.
Would
would
there
be
this
accelerated
sort
of
devaluation
of
that
soakin
if
the
prime
or
significant
amount
of
its
value
was
derived
from
the
fact
that
it
was
a
collateral
fight,
but
it's
no
longer
a
good
collateral
type?
Would
that
cause
it
to
you
nosedive
even
more
quickly
than
what
have
other
ones
well,.
B
That
all
if
your
risk
management
is
not
particularly
good,
yes
you'll,
contribute
towards
that,
and
there
is
a
in
the
traditional
to
the
finance
space.
You've
got
a
lot
of
small
cap
companies
that
go
to
banks
for
financing
and
they
do
use
their
own
stock
as
collateral.
But
you
know
going
to
keep
in
mind
that
you
might
you
might
find
a
small
cap.
A
B
B
Yes,
I
mean
that's
the
same,
let's
say,
for
instance,
we
get
someone.
That
is,
let's
pretend
for
a
moment
that
I
would
include
someone
that
has
a
10
million
market
cap
getting
excluded
and
bumps
up
to
City.
Now,
I
consider
this
for
a
moment
as
well.
We've
got
a
debt
ceiling
on
this
and
you
might
have
let
say,
exposure
to
I
wouldn't
put
past
four
or
five
percent
of
market
cap,
so
that
bump
would
basically
turn
into
not
15
percent
of
our
original
use
of
the
debt-ceiling.
B
The
point
there
is
to
always
keep
the
assessment
going
to
always
keep
the
analysis
on
the
token
and
the
organization
going.
That
is,
that
is
the
competitive
edge.
That
is
where
the
thing
works
is
that
very
much
like
our
governance
or
the
the
the
executive
components
is
based
on
continuous
approval
voting,
so
should
the
risk
management
should
be
a
continuous
assessment
of
not
only
the
token
construct,
but
also
the
organization
that
represents
it.
We
did
say.
A
Sorry
is
because
it
sounds
like
an
enormous
ly:
complex
and
fast-moving.
Sort
of
algorithm
needs
to
be
balanced,
but
I
think
that
that's
because
I
made
the
mistake
of
thinking
that
we're
only
going
to
be
supporting
ERC
20s
in
in
the
future
and
I
think
that
that's
a
very
you
know
kind
of
unpredictable,
very
fast-moving,
a
kind
of
a
rational
market,
but
with
a
focus
on
real
world
assets.
And
it's
organization
of
you
know
real
life.
That
kind
of
correlation
risk
obviously
goes
way
down
and
I
kind
of
you.
B
Would
think
your
Thanksgivings
right,
you
think,
here's
another
here's
another
little
sort
of
thought
process
to
keep
in
mind.
There's
something.
Obviously,
when
you
have
a
look
at
I,
don't
know
how
much
you
guys
would
know,
but
if
you
look
at
the
futures
market
there's
something
called
the
basis
risk
which
effectively
says
well,
what's
the?
What
is
the
the
the
true
difference
between
the
futures
that
you're
using
to
hedge
or
unlined
commodity?
Now
there
is
this
little
risk
and
that
risk
is
representative
of
how
people
are
trading
in
the
market.
Now
the
same
thing
applies.
B
You
know
to
us.
We
have
to
have
a
look
at
what
is
the
basis
risk
between
the
token
that
represents
an
organization
and
the
organization
itself.
You
might
find
that,
like
an
exchange-traded
fund,
even
though
by
construct,
you
might
have
exchange-traded
fund
move
away
from
its
it's
fair
value
during
the
day,
but
ultimately
that
fair
value
needs
to
be
realigned
to
the
price
by
the
end
of
the
day.
In
our
sense,
if
we
don't
have
that
mechanism
place
in
the
securities
market,
you
might
find
the
token
is
completely
dislocated
from
the
organization's
value.
B
You
might
find
the
token
is
all
over
the
place.
It
may
have
more
volatility
in
the
actual
organization's
security
hat.
So
let's
say,
for
instance,
we
do
something
like
tokenized,
Walmart
you'll
have
a
token
that
represents
Walmart
and
you
have
a
security
that
represents
Walmart.
Theoretically,
they
should
be
in
lockstep,
but
the
problem
being
is
you've.
B
The
other
version
is
folks,
have
scrip
and
they
just
use
that
scrip
in
and
turn
it
into
talk,
and
they
might
legitimate
have
token,
or
they
may
basically
just
I,
store
it
and
and
and
produce
new
tokens
on
it
it
it
depends,
but
that
method
produces
a
little
bit
of
basis
risk
and
that
you
gotta
consider
as
well,
because
this
is
not.
This
is
not
it's
not
always
straightforward.
It's
always
the
case
of
oh
look,
you
know,
Walmart
is
going
to
be.
B
You
know
a
wonderfully
uncorrelated
as
it
compares
to
ERC
twenty
turtles,
but
as
soon
as
you
turn
it
into
the
ERC
20
token,
you
might
find
that
captures
a
lot
of
correlation
of
the
ERC
20
environmental
industry.
The
hope
is
that
it
doesn't
hope
is
that
what
it
brings
in
will
start
creating
assets
that
are
completely
different
from
the
g20
environment.
A
Right
we're
coming
up
on
nine
minutes
left
to
the
top
of
the
hour.
If
anybody
has
any
questions
about
anything
at
all,
please
feel
free
to
ask
one
of
the
things
I
I
think.
Maybe
this
touches
a
bit
on
what
you
were
just
discussing
Stephen,
but
it's
a
question
that
comes
up
in
every
so
often
enough
is
well
making
down
ever
consider
having
an
asset
class
of
tokens
like
the
ones
that
digit
I
miss
are
dy/dx,
did
X
offer
so
a
short,
tough
boost.
B
A
B
Matter
what
token
is
considered
and
always
boils
down
to
recourse?
Ultimately,
you
can
have
a
a
derivative
based
token.
You
could
have
really
a
compound
token
of
others.
As
I
said
previously,
you
could
you
could
construct
a
fungible
token
from
non
fungible
tokens
and
have
that
submitted
as
a
collateral
type.
Ultimately,
is
the
acceptance
of
it
boils
down
to
the
ability
to
recourse
all
the
way
back
to
your
final
satisfaction
of
of
an
event,
some
sort
of
liquidation
in
to
interferes
or
some
sort
of
liquidation
into
a
mole
liquid
token.
No.
A
Because
that
makes
sense,
I
think
of
pulling
into
the
same
trap.
So
a
lot
of
people
fall
into
when
they
think
of
what
the
collateral
types
and
that
is,
does
X
work
and
generally,
the
answer
is
yes:
if
there's
liquidity,
if
there's
recourse
and
if
there
is
reasonable
risk
parameters
placed
on
it,
of
course,
anything
can
basically
come
in
its
yeah.
Yes,
but
it's
after
it
passes
all
those
hoops
I
guess,
what's
the
utility
of
actually
having
that
token
in
the
asset
class,
because
it.
B
B
B
In
other
words,
your
ability
to
go
and
sell
something
into
an
open
market,
which
then
moves
you
to
think
clearly
about
what
an
open
market
is
sorry,
what
an
open
market
is
and
at
the
same
time,
how
you
define
one
and
what
type
of
market
you
want
to
be
able
to
interact
with.
In
other
words,
what's
the
liquidity
is
liquidity
real?
B
Where
does
it
come
from
the
distribution
of
players
and
the
distribution
of
incentives
in
a
particular
market
needs
to
be
diverse
enough
to
make
that
liquidity
sustainable,
though
these
aspects
of
being
looked
at
in
the
traditional
market
for
the
longest
time
now
in
developing
markets,
creating
markets
and
sustaining
markets
and
your
market
make?
Is
there
you
know
being
in
specialists?
Or
you
know,
specialists
on
a
stock
or
specialists
on
our
security
or
just
a
general
market
maker
you'll
find
that
they
are
very
cognizant
of
the
distribution
of
players
in
the
market.
B
As
an
example,
you
find
that
a
lot
of
movement
is
going
towards
your
sort
of
algorithmic,
well,
actually
I'd
say
predominantly
70.
80
percent
of
the
the
action
in
additional
market
is
algorithmic,
so
that
means
you've
actually
reduced
the
number
of
players
that
you
have
in
a
theoretical
sense,
because
you
don't
have
a
disbursement
of
people
all
phoning
in
the
course
now
being
sort
of
abstracted
into
teams
that
represent
specific
algorithms
or
teams
that
that
represent
specific
Robo
advisors
and
those
Robo
advisors
and
algorithms.
B
They
didn't
have
the
reduction
in
latency
to
pick
up
the
fact
that
the
market
was
crashing
and
they
should.
Maybe
you
know
flag
a
human
being
to
say.
Maybe
you
just
need
to
ride
this
one
out.
None
of
that
ability
was
dead.
It
was
just
a
massive
supply
that
came
in
crushed
market
for
a
second
and
then
it
revived
and
someone
switched
off
machine
and
actually
bought
stuff
back.
B
No,
no,
but
it's
fun,
yeah!
That's
the
thing:
if
it's
not
fun,
it's
not
we're
doing.
The
the
one
aspect
here
is
that
you
know
there's
so
many
smart
guys
out
there
and
what
I'm
trying
to
say
is
that
there
is
always
if
you
look
carefully
enough
a
place
to
apply
your
skills,
there
is
no
avenue
in
the
risk
spectrum
that
says.
Oh,
this
is
just
for
risk
I.
These
are
just
for
mathematicians.
This
is
just
for
this.
A
Well,
yeah
it
is,
it
is
gonna,
be
a
lot
of
fun
actually
and
I'm.
Looking
forward
to
having
some
more
discussions
about
how
we
begin
to
reach
out
to
the
communities
are
aggregating.
Some
of
the
ideas
and
interest
that
we've
seen
from
some
of
our
members
and
attracting
more
I
think
that's
would
be
a
cool
discussion
for
a
future
meeting
at
pier
CLE
all
right.
We
have
two
minutes
that
are
there
any
final
thoughts
you
wanted
to
leave
us
with?
There
should
be
wrapping
up.