00:00:08
hi I'm Ken Smithers professor here at
00:00:11
the Wharton School and I'm getting give
00:00:13
you a quick overview
00:00:15
of the new social security policy
00:00:17
simulator that's available at the pen
00:00:19
Wharton budget model group and you can
00:00:23
see from the screen here is that you
00:00:26
have a choice of various graphs that you
00:00:29
can be looking at things like trust fund
00:00:31
reserves non-interest surplus which the
00:00:34
difference between benefits paid and
00:00:36
taxes income receive Social Security
00:00:39
taxes benefits interest income lots of
00:00:43
other varieties and we also compare our
00:00:46
results against those from the Social
00:00:48
Security Administration and the
00:00:50
Congressional Budget Office and first
00:00:53
also BP before we do even think about
00:00:55
any policy changes you can see that our
00:00:58
model projects that we will exhaust the
00:01:04
trust fund exhausting day will be about
00:01:07
three and a half to four years earlier
00:01:09
than the Social Security trustees
00:01:12
project in particular we see our
00:01:17
deficits happening around 20 31 and what
00:01:22
come 20 31 we're seeing projecting that
00:01:26
will actually have a non interest
00:01:28
surplus of about 350 billion dollars in
00:01:33
today's dollars eventually growing to
00:01:35
over 1.2 trillion dollars per every
00:01:39
single year again in today's dollars an
00:01:43
hour 75 years shortfall is about a third
00:01:47
larger than what you see from the Social
00:01:49
Security Administration and to
00:01:51
understand why you can look at the
00:01:54
social security tax revenue and we have
00:01:57
a more pessimistic projection of tax
00:01:59
revenue coming in over time and that the
00:02:01
reason behind that is that our model
00:02:05
starts with census level data and we
00:02:07
construct these transition rules based
00:02:10
on various big data sets across many
00:02:12
different key attributes age race
00:02:15
marriage divorce number of kids
00:02:17
education and many other attributes
00:02:20
across a lot of population subgroups so
00:02:22
we're really working kind of bottom-up
00:02:24
and as a result of that we can see how
00:02:26
productivity in the economy
00:02:28
is changing as you have these large
00:02:30
labor force composition changes as
00:02:33
people are going into retirement they're
00:02:37
being replaced by younger people and the
00:02:40
average productivity per worker for
00:02:41
example goes down as a result of that
00:02:43
our benefits also go down a little bit
00:02:46
relative social security but there are
00:02:50
tax revenue is is is the main difference
00:02:54
in our numbers line up a little bit
00:02:56
closer to the cbo's own model on this
00:03:01
all right so now let's consider some
00:03:02
different policy changes and we have six
00:03:05
options to pick from here increase in
00:03:07
the payroll tax increase in taxable
00:03:10
maximum and lots of other options
00:03:12
there's altogether there's six options
00:03:16
each has four different combinations
00:03:19
associated with it so there's actually
00:03:21
four to the sixth power different
00:03:23
combinations 4096 policy combinations
00:03:26
that you can run in our model you know
00:03:29
each one takes about 30 minutes than our
00:03:31
to run so you might be wondering how
00:03:33
you're getting instant feedback as you
00:03:36
move things around and different policy
00:03:38
changes and we the reason is we're using
00:03:40
cloud computing to do all the
00:03:42
computations ahead of time and so that
00:03:44
gives you instant feedback let's
00:03:47
consider a couple policy changes that
00:03:49
policy makers for years have been kind
00:03:51
of talking about let's for example talk
00:03:54
about raising the normal retirement age
00:03:56
this one down here and we have a little
00:03:58
tool tip here with it basically says
00:04:00
what the current law is in particular
00:04:04
the retirement age is going up to age 67
00:04:07
roughly about two months per year and if
00:04:10
you want more info it can give you even
00:04:11
more expanded info information and so
00:04:14
this would be a policy change to
00:04:16
continue that increase if you wanted to
00:04:18
do that say even to age 70 and now when
00:04:21
we do that you'll notice here that the
00:04:25
impact on the trust fund exhaustion date
00:04:28
from increasing the retirement age it's
00:04:31
basically less than one year it's a very
00:04:33
small change and by the way if you
00:04:36
wanted to you can even kind of zoom in
00:04:38
on the particular years and kind of
00:04:41
it get more information you can you can
00:04:43
actually look at particular yours the
00:04:46
values and so forth and it doesn't have
00:04:48
a big impact however over time increase
00:04:53
in retirement age does have a much
00:04:55
bigger impact on a non interest surplus
00:04:57
the big deficit over time goes down and
00:05:02
then what's going on here is essentially
00:05:05
that we're increasing the retirement age
00:05:08
is is simply basically too late to solve
00:05:11
the problem at least for the trust fund
00:05:13
it can be part of a longer term solution
00:05:16
but it just simply faces in too slowly
00:05:20
over a time and then we have some other
00:05:24
things that you can consider here when
00:05:27
is a progressive benefit reduction and
00:05:30
you hear it explains the tooltip
00:05:34
explains how benefits are currently
00:05:35
calculated in particular the government
00:05:38
you currently calculates your benefits
00:05:43
equal to 90 percent replacement rate up
00:05:46
to a certain value called a bend point
00:05:49
then thirty two percent and then fifteen
00:05:51
percent and so this is a policy where
00:05:54
you keep the ninety percent replacement
00:05:57
rate for poorer households constant but
00:05:59
you decrease the replacement rate on
00:06:01
Richard households and you can see the
00:06:03
impact on the trust fund is actually
00:06:06
also a fairly small and the current
00:06:11
policies are blue line the red policy is
00:06:13
kind of the new line and it barely kind
00:06:16
of moves the trust fund exhausting a
00:06:20
date and saying the intuition behind
00:06:24
that is again it just is a bit too late
00:06:27
they have an impact on things that
00:06:30
phases into slowly however again over
00:06:33
time it has a much bigger impact and
00:06:36
then you can start to do interactions
00:06:37
like you know suppose I want to do a
00:06:40
progressive benefit reduction increase
00:06:42
the normal retirement age and you can
00:06:44
now start to see that it has a much
00:06:45
bigger impact over time in terms of the
00:06:48
trust fund reserves themselves again not
00:06:51
a huge impact at least
00:06:54
in the short run okay so now let's
00:06:57
consider you know other things I mean
00:07:00
you can a lot of people have argued for
00:07:02
like increase in the taxable maximum
00:07:04
right now your payroll is being taxed up
00:07:08
to 118 thousand dollars 500 per year a
00:07:12
12 point four percent tax rate and so
00:07:15
some people have argued that maybe we
00:07:17
should levy that tax over our larger
00:07:19
income base and what we can see is for
00:07:22
example as we go to the say increase
00:07:27
that $250,000 it actually has an impact
00:07:31
on on the trust fund as we go to 400,000
00:07:34
as a slightly bigger impact not much of
00:07:36
a difference going from 250 to 400
00:07:39
thousand households there's just not
00:07:41
enough households there and if you go
00:07:43
above 400,000 house of dollars it has
00:07:46
almost no impact going forward but again
00:07:49
we get some life but you know over the
00:07:52
long term it starts like you know I get
00:07:55
a little bit bigger this is one of those
00:07:56
policies as a much bigger impact on the
00:07:59
short run because it's immediately
00:08:00
happening then but the relative to the
00:08:04
long term and so you can now start to
00:08:06
think about different combinations I
00:08:07
mean there's a 4096 combinations here
00:08:11
that for you'll explore I I haven't even
00:08:13
explored the every single accommodation
00:08:15
and so you can see the trade-offs
00:08:18
yourself especially between the long run
00:08:20
and the short run by giving your own
00:08:23
combinations a try thank you
