How-To and Q&A User Webinar - A deep dive into Historical, Traditional Monte Carlo, and Regime-Based Monte Carlo simulations - January 2022

hello everyone thanks for joining we are going to give start in a few minutes as

0:18

we have uh folks getting in on here oh and perfect there's eric

0:25

morning derek hello i guess afternoon at this point

0:38

seeing some more folks coming in

0:44

okay let's give it a few more seconds

0:52

all right okay well welcome everyone to our first uh user webinar at income lab of the

0:59

year happy new year to our existing users to our new users welcome

1:05

to our monthly webinars we are excited to bring our two experts to the

1:11

to the panel uh justin our co-founder and derek tharp our uh senior advisor

1:16

and um you know head practitioner i say who always gives us great feedback on you know how advisers are using and

1:23

talking about income lab um with our webinar this month we are going to take a deep dive into you know our um

1:30

analysis method so historical analysis traditional monte carlo regime-based monte carlo and really help users have a

1:37

better understanding of what those are and how that really plays into income lab um for those of you who this is your

1:43

first webinar uh we will also hold a q a session at the end um so you'll see in

1:49

um our zoom webinar space here that there's a section for q a go ahead and drop any of your questions into this um

1:56

section you'll also be able to like other people's questions so if someone has a question that you also want

2:02

answered you can like it and move it up in the queue and then at the end of the presentation i will kind of just work

2:07

through the queue and and help answer any questions um outside of that we will also record the webinar so after

2:14

tomorrow um if you want um to kind of re-watch the webinar you'll also be able to do so as well

2:21

outside of that guys welcome um i will turn it over justin to you and derek and

2:26

i'll be back when uh for q a okay great thanks and happy new year

2:31

everybody um share my screen here so today's

2:37

webinar um and for those of you who are new usually we try to do about half an hour of um presentation

2:44

and then open it up for um q a for up to half an hour um so today's uh topic

2:52

it really dives into some of the some of the background or some of the underlying parts of the income lab

2:59

platform um and so some i think some of what we'll talk about today and i think you know maybe

3:05

even derek's um commentary on this will be you know how much of this to kind of

3:10

expose to clients how to talk about it and things like that but a lot of it is is really about kind of your settings

3:17

your choices in developing plans and so on so we call it what's your

3:22

model of the world because really let me make sure yep

3:31

we want to think about this as when we're building a plan you know what kinds of inputs and analytics go into it right so

3:38

our goal is to get this retirement income plan um our goal is to get

3:43

um you know ideas measurements concepts pictures um that we can use to communicate with

3:50

clients and to guide them through retirement or pre-retirement um but a

3:55

lot goes into that right so we have what i'm calling household particulars which is really all of the um

4:02

the special things about a household it could be their age their longevity preferences their risk tolerance their

4:07

the types of accounts they have how much money how it's invested um when they're taking social security all of those you

4:13

know particulars but we also need these models right these other things about how we think

4:19

the world works um so i've divided them here into your market model your economic model your actuarial model

4:27

and so on right and i keep saying that word model because a really important thing to

4:32

remember in any kind of modeling is our model of the world is not the world

4:37

right so quick uh art history lesson here probably many of you have seen this before right so this is rename agreed this is

4:44

not a pipe um it's a representation of a pipe right so you can't pack this with tobacco

4:52

same thing goes for our models that doesn't mean we're gonna we're we're gonna you know sort of give up throw our

4:57

hands up on modeling the world no we we want to be as accurate as realistic as possible but

5:04

there is simply no getting away from this problem right that we can't predict

5:10

the future and that we're always making decisions and and um

5:15

and plans within a model of the world right so

5:20

this uh this fact is actually it's very relevant right this is not just some kind of weird esoteric academic concept

5:28

model risk is very real um there's been a lot of written about the 2008 2009 financial crisis

5:34

talking about model risk right it's this risk of loss resulting from insufficiently accurate models when

5:41

they're used to make decisions right and we can kind of think about this for retirement income planning as well as

5:47

the risk of an adverse outcome due to significant differences between

5:52

what we're modeling and the actual outcomes and i i would say there's two really important things here

5:58

one is you actually have to have an a risk of an adverse outcome right so it has there has to be some some so what

6:06

about it for there to be real risk um and uh the other is

6:11

you that there have to there has to be something where we're actually using this to make decisions right where it's sort of somewhat in our control um so

6:20

for you know the the more general case and the example i have of the financial crisis um

6:25

again lots of articles about kind of mistakes and modeling credit derivatives um you know sort of assuming things

6:32

couldn't be as as as they were for our purposes in retirement income

6:37

planning it might be something like using capital market assumptions that are significantly higher or lower

6:43

or maybe return inflation assumptions that are higher or lower than the household might actually experience

6:50

um but the real takeaway is all models have model risk so there's really no way to completely um get rid of of model

6:58

risk so that's kind of the overarching kind of philosophical underpinnings of this discussion

7:04

um here are some ways that you could some kind of uh you

7:09

know statistical approaches that you could use for your economic model or your market

7:15

model and by economic model here i'm mostly talking about inflation um so uh and for market we're talking about

7:22

you know investment returns you could use straight-line appreciation this is um you know still commonly used

7:29

anytime you're using spreadsheets it's sort of the uh you know the excel present value future value payment

7:34

rate um sorts of calculations or it's your financial calculator um traditional

7:40

monte carlo and and then in income lab we also have regime based monte carlo and historical

7:46

um it's actually interesting so straight line appreciation it's it's pretty easy to

7:52

you know kind of um uh you know look down on this and i think there are reasons to but even in income

7:59

lab we use straight line appreciation where we where we have to anywhere where something more complex or more realistic

8:05

would be confusing or misleading so here's an example you know if you're looking at the the income sourcing chart

8:13

uh in with in the nominal view so including inflation so looking at future dollars we have to show something right

8:20

and so we use a straight line appreciation uh for inflation in this case using anything else would be very

8:26

confusing so what if we used you know varying inflation which is what we use in every other part

8:32

of the of the application where we can well you might see this line kind of going up and down maybe there's deflation some years

8:38

that's just unnecessarily complex to explain to a client right so it's very reasonable to use straight line

8:44

appreciation in certain cases but obviously straight-line appreciation for um households that actually have

8:50

investments would be problematic because we know that investment returns can can vary and are

8:56

not identical every year and so that's really where

9:01

more complex analysis types came into financial planning most famously monte carlo approaches um

9:10

an income lab offers both historical and monte carlo what we call analysis

9:18

types and this is one of the places where um you know you may want to explore

9:23

what uh approach you prefer to use so you know we don't

9:29

we have maybe personal opinions but we certainly don't have kind of a um an overall opinion on which one you

9:35

should use there are pluses and minuses that we're going to go over um here

9:42

so monte carlo um is is just a way of of kind of

9:49

creating return sequences lots of different possible approaches we use a thousand scenarios in our monte

9:56

carlo analyses um in order to show that look returns could be um could vary year

10:03

by year month by month so in income lab we we do things monthly whenever we can and so we're talking about monthly returns here

10:09

um on the monte carlo side we offer both the traditional monte carlo which is similar to what you'd find in a lot of

10:15

other applications where you're using one set of capital market assumptions for all the months in the plan

10:22

and we have a regime based monte carlo that uses two sets of capital market assumptions one for the first number of

10:28

years and that's a setting that you can choose we use our default is for the first 10 years and the other set for the

10:34

remainder of the plan however that however long that is if the plan is less than 10 years in in that case you

10:40

wouldn't be using those at all if it were 30 years it'd be for for the uh you know the 20 years that

10:45

follow the first 10. um for historical analysis we use index return sequences so actual

10:52

inflation-adjusted return sequences um and actually actual inflation sequences

10:58

as well um for the model let's um let's take that second one

11:04

first because i think there's a lot to talk about here um because there aren't many applications that that let you use

11:10

historical analysis um so i think it's worth worth discussing some things about it

11:16

um so as with with any kind of modeling there's advantages and disadvantages probably the main advantages are um

11:23

we're only going to be looking at scenarios we know can actually happen because they have happened

11:28

so i sometimes joke that the term historical data is redundant the only kind of data we have is historical data

11:37

another positive thing here is you know people will sometimes ring their hands about monte carlo analysis because it's

11:43

really hard to capture all of the complexity of of how investment returns really behave

11:52

so the example that i'd like to use here is um you know people joke that uh

11:57

that in in times of uh you know market turmoil the only thing that goes up is correlations

12:03

so when you're doing monte carlo analysis you typically provide one set of correlations among asset classes but

12:10

those correlations are the same no matter what when you use historical analysis you're able to to include in

12:16

that range of possible scenarios scenarios where correlations went up right where correlations themselves

12:22

varied or where standard deviations varied so these kind of second and third level statistics are

12:28

they're just baked into historical analysis but they would be very difficult to put in monte carlo analysis

12:34

and in practice they they never are in financial planning another advantage which we're going to

12:39

talk about right now is that you're able to bring in economic context so you're able to

12:45

talk about economic context so there's no sense in which you could do you know economic context using monte carlo right

12:52

randomizing my economic context wouldn't be very helpful but there are disadvantages right um so

12:58

we only have so much history we use 150 years of history and so we're not considering return

13:04

sequences that haven't occurred but could that also mean certain asset classes

13:09

maybe the newer asset classes or more esoteric asset classes you don't have as much data for

13:15

and sometimes people talk about the fact that those return sequences you're using they have some overlap right so you know

13:23

january of 1960 will appear in all sorts of uh return sequences

13:28

um so let's talk about um economic context briefly and why that's that's helpful in an income lab

13:36

no matter what analysis type you're using you'll always have access to

13:41

the historical analysis tab so even if you're using monte carlo traditional or regime based we will still run

13:50

some historical analysis in order to be able to show you this picture so this picture shows you

13:55

um for this particular plan with all its idiosyncrasies the social security it has the timing maybe it's using the

14:02

retirement smile everything what someone following this plan

14:08

could have spent could have had an income at every point in recorded history for

14:14

which we know that right so for any any situation where we where we know what their experience would have been so we

14:20

can look back at the 1960s and say hey if someone was living out a 30-year retirement after you know 1965 what's

14:28

what could they have spent and we know that because we know what what happened after 1965 right

14:34

well clearly that just just getting that number requires historical data but also we shade this graph

14:41

to be blue and gray blue being periods that are most like today in economic terms gray being those

14:47

least like today and in order to do that shading we bring in we consume a lot of data at income

14:53

lab and so we bring in all sorts of economic data to kind of produce that shading for you so you can

14:58

see at the bottom we're using inflation and treasury rates and unemployment and cape and

15:03

average investor equity allocation and the yield curve and so on to kind of build this composite picture of hey

15:10

you know what do all these periods look like and which ones were most like today so because even if you're using monte

15:17

carlo you may want to talk about this with clients we always do that for you even if you're using monte carlo

15:24

so how could you use economic contacts um there's at least two ways one is you could use it to to kind of

15:31

actually affect your income advice to people so when income risk seems particularly high or low tilting your

15:38

advice up or down um i guess actually the reverse of that so down when it's high up when it's low another would be

15:45

um to enrich your client communication to helping clients kind of understand the connections or lack thereof between

15:53

economics and retirement income and of course you could do both as well

15:59

if you were allowing it to affect the advice you give you'd be basically relying on

16:05

this the sort of patterns that we see from certain economic factors to income

16:13

outcomes of a certain sort so this is the um probably the most commonly discussed

16:19

uh economic indicator with respect to retirement income which is cape so cyclically adjusted

16:26

p e and i've just picked out the relationship between cape

16:31

and um 20-year withdrawal rate so very simplified just saying okay we're gonna

16:38

we're gonna withdraw an inflation-adjusted amount for 20 years you know how much could i take and

16:45

again using history because that's that's the only way to do it using economic context and you can see there's

16:51

a there's a trend as cape is higher historically people have been able to afford lower um

16:57

lower withdrawal rates right so you'd be if you allow economic context to to be part of your income advice this is this

17:04

is what you're leveraging right um you can also use it to help in

17:09

communication so um i know recently um some advisors have said you know i've

17:14

had clients asking me hey um you know i'm i'm really worried right

17:19

things have been so good in markets for so long but you know geez it seems like there's a lot of

17:25

kind of turmoil in the world and they start thinking about things like the great depression right so you can use this chart um maybe

17:33

you you pick out the wall street crash of of 1929 and say well look this is

17:38

what happened during the great depression um and if we apply your plan to it this is what um what it would look

17:44

like um and so we we've taken that into account here right and you can see in this case the black line which is their

17:50

proposed income is actually below the worst that happened during the great depression

17:56

and so you know could the future be worse than the past of course but in a way you've kind of

18:01

the plan you're adopting here is uh is is great depression proof uh at least as

18:06

it played out historically so that can be really useful for providing contacts for clients where you know statistical

18:14

measures are are not as insightful for them right talking about things like probability of success or

18:20

monte carlo analysis is not as something that you can connect with as easily

18:27

um turning now to money carlo analysis um advantages and disadvantages and i think depending on your opinions uh

18:33

you know sometimes uh you could say each of these is both an advantage and a disadvantage so you can explore

18:39

scenarios that have never occurred but of course you might be exploring extreme outliers that we might think are maybe

18:47

will never occur or would we are very unlikely to occur you can express opinions um that's

18:52

something you can't do with with history but you can't connect economic context as we just discussed you can cover asset

18:59

classes with short histories but of course you have to have assumptions about how those will behave

19:05

but you can't really include all realistic market behavior factors like changing

19:10

correlations right so pluses and minuses and there's lots written about this derrick has written quite a bit about

19:16

pluses and minuses of monte carlo so i may have some some comments there

19:21

um within income lab we have two approaches as we've already said we have one

19:27

with one set of capital market assumptions and we have defaults for you there we try very hard to make our

19:32

default capital market assumptions um very formulaic so they're produced um

19:38

formulaically using uh historical averages so this is not you know income lab's

19:44

opinion about what your capital market assumption should be we just want you to have very reasonable defaults available

19:50

for use um and this is you know this this uh production of defaults is very similar

19:57

to what you might find in other monte carlo based analyses for regime based

20:03

um this is a place where to my knowledge we're unique that we have regime based monte carlo

20:09

and again you have a near term and a long term are defaults again they're formulaic but it's it's

20:14

it's a fairly sophisticated approach what we do is we take um history

20:20

we take we create rolling 30-year periods and we exclude two-thirds of

20:25

them the ones that are at least like today and then we do averages so averages for

20:31

the first 10 years averages for the following 20 years to create near-term and long-term

20:37

and so you'll see what typically happens there is you know maybe near-term returns are

20:42

lower um and then long-term returns are higher right in a sort of reversion to the mean type pattern

20:48

again though this is not the income lab opinion or anything it's just a formulaic way to uh to produce defaults

20:57

one thing that this can help with you know depending on where you source your capital market assumptions they're not

21:04

they're not always made for retirement income planning right they might be something more like five or ten year

21:09

capital market assumptions for investment planning um so it's important to think about kind of what the time

21:15

scale the intended time scale is for those capital market assumptions because

21:20

you know different plans are different length but it's very common to be planning for 20 30 40 years

21:26

and so maybe a shorter term or medium-term capital market assumption set

21:32

you know applying that to 30 years may be maybe problematic so uh within the platform

21:39

within income lab if you go into your settings and then click capital market assumptions this this menu is just

21:46

reworked slightly to make it a little clearer you'll find the capital market market assumptions section if you are the firm

21:53

admin you have an option to allow other advisors in the firm to edit their

21:59

own capital market assumptions you could uncheck this if you have firm assumptions that you are applying

22:04

everywhere and then you have three sets of capital market assumptions traditional

22:10

regime based near term and regime based long term and for each of these you can either use the defaults

22:16

which we provide or change this toggle to custom and you're able to uh to input your own

22:24

again for regime based monte carlo you have two sets uh one for the beginning of the plan one for after that

22:30

um and you can under the regime based near term tab if

22:35

you hit custom you can change how long the near term assumptions apply

22:41

so our default is for 10 years but you can you can adjust that and long-term will apply for any any period

22:47

after that okay so that's a lot about kind of where

22:54

these analysis types come from the kind of inputs but really what's the purpose of

23:01

this what are we after right it's it's not you know we're just really interested in math and we want to talk about math with our clients or something

23:07

right there's a there's a purpose of this which is to provide good advice so one answer to that question what's

23:13

the purpose of analysis types or capital market assumptions is well we want to get a picture of

23:18

the future that clients might be living through right we want to we want to paint

23:23

a kind of the landscape that they're traveling through as as realistically as we possibly can and so we take

23:29

the analysis types the cmas actuarial data which here would just be giving us

23:35

the plan length and all those household particulars and we want to figure out you know what kind of income risk do

23:40

income levels have and so on so we're we're producing again not

23:46

this is this is still kind of um you know deep math and things we probably wouldn't

23:52

show clients but we're producing this picture of the world where okay there are different income levels that they

23:57

might be able to access in their retirement and each has its own level of risk and we're estimating what levels of

24:04

risk are there right so so all of that randomization or using history and and

24:10

you know standard deviations and so on that the purpose of that is to figure out a picture of income and income risk

24:17

and ultimately to provide as much as possible data that is you know at the right level

24:23

of abstraction as uh derek has talked about a level where

24:29

you know our clients can can access kind of a connection with that as easily as possible so for example the

24:36

proposed income and the income adjustment plan all stated in dollar terms right something that's very easy

24:42

to connect with well everything we just talked about with analysis types and um you know the pluses and minuses of

24:49

each and so on um those are behind these numbers right but these numbers are are now simplified and

24:56

presented in a straightforward way also um they provide things like the the plan

25:03

tests to the you know what what could this plan feel like if followed over

25:08

time right um might we expect to spend more or less than expected um how often might we have

25:16

um increases or decreases and so on so everything we just went over is behind

25:22

this um and and it's important to keep in mind that that's that's kind of the purpose of of making choices about

25:29

analysis types capital market assumptions and so on is is to get to these outputs um that that can be useful

25:36

for for client communication and for good planning so that's the the presentation i have so

25:43

far i imagine that there are um a bunch of questions but uh before we get there maybe if derek if you have any

25:49

kind of comments um on on making these kinds of decisions or talking with clients

25:56

yeah no i think for me just having you know the ability to dive deeper if i need to um and

26:03

understand what i'm doing as an advisor and why i'm doing that um is really powerful you know using a

26:08

tool like regime-based monte carlo i think that's a really neat thing to do

26:14

but then in practice am i actually talking to most clients about that no probably not

26:19

it is you know really finding the right situation i can go deeper if i need to with the right type of client that has

26:26

questions about that but for most of my clients this isn't what's on their mind i mean the kind of the guard rails presentation

26:32

that you can see there in terms of the current spending level when the adjustments are going to occur

26:38

you know that's really where our conversation is focused and then making sure

26:43

we that lines up in terms of the long-term income outcomes somebody could

26:49

experience lines up with what they actually want so you know i'd say today's presentation is much more for the

26:56

advisor um and how we do what we do and do it well

27:01

um even though it may not actually make it into some of the conversation although i think justin provided a

27:06

number of good ways that you could have some conversations around specific issues and education

27:12

or clients on those but yeah for for me the most part uh this is living in the background

27:22

all right guys um let me got a few questions in here um

27:29

okay so first one um you know how many asset classes [Music]

27:35

i guess if you guys can comment on how many asset classes should be used for capital market assumptions the question is if you could comment on whether

27:41

including too many asset classes increases the likelihood of model risk or if it's better to use a few common

27:48

major asset classes you know like lap large cap mid cap small cap etc

27:57

so i can answer the the first section first set there so again this has been slightly

28:02

redesigned so i believe there are 15 asset classes

28:08

right now if you include inflation we are actually in the process of expanding that so there will be a um an

28:15

all-cap value and growth as well as three mid caps some mid cap

28:21

plus value and growth um i believe we have a short-term bond coming in

28:27

and there may be one other commodities i think so those

28:32

that's that's in the process as many of you have asked for more asset classes here

28:38

um on the second comment on whether it increases model risk or not um

28:44

the derek might have ideas on this i do think on historical we're very

28:50

careful to use asset classes where we think we have enough history that historical analysis

28:55

still works and that's maybe frustrating to to some users who would like to see

29:01

newer asset classes that doesn't mean we never would include those but we are as we expand these we're trying to make

29:08

sure everything works in all of the available analysis types and so one worry i would have is if you don't have enough history to

29:15

to paint a a reasonable picture um i think on on the second point it sort

29:22

of depends on on you know what sorts of portfolios you're putting together you can even do a

29:27

little bit of a b testing and see how much of an effect it has on on the you know the outcomes that you're

29:33

like i said you know model risk is really there if it affects your your decisions right and and if it could have a an adverse outcome so if it's a if the

29:40

difference in income is you know 10 bucks or something probably it's not a big model risk um i don't know derek i

29:46

know you've done a lot of work on monte carlo if you have thoughts there it's an interesting question i'm kind of

29:52

just answering from intuition and not not necessarily as well grounded in a

29:58

research response as i normally would like to be but i think my initial reaction is that it does kind of depend

30:05

on what sort of portfolios you're building you know if you're putting in a lot of different asset classes but

30:10

things are pretty spread out across them they're not really concentrated you know then i don't think um i would be as

30:16

worried but if you you know you're modeling or using portfolios such that you know there is

30:22

this really broken out granular asset class selection and then you're

30:28

putting you know i'm just making up an example let's say you're putting 50 in small cap value or something to me then

30:34

you might elevate your risk of just getting some historical anomaly and extra putting too much weight

30:41

on that but if you you know have a pretty well and diversified across the asset classes

30:46

then i don't so i do think it is how you're using it the more heavily weighted you are in a small

30:53

number of granular categories the more concerned i would potentially be

30:59

um but yeah that comes down to more how you use it and this is just a quick one in the chat

31:05

is there a life insurance asset class

31:10

no there's not you can put life insurance in um as you know kind of a protection um or a

31:17

cache value um but but no all the the the asset classes are just

31:22

well i'd switch what you're looking at so you can see them here um yeah and so the next one uh this

31:29

one's got a few upvotes so um how does income lab account for historic low interest rates at the beginning of the

31:35

first regime um and do rising interest rates um show as negative

31:41

returns in bonds i.e do you illustrate sequence risk and bonds yeah so um we have a

31:48

i can show you kind of these are the um well i'm not in the production environment so hopefully these are the updated numbers

31:55

but if they're not they're they're close um yeah it looks like they're pretty close um so you can see

32:01

again this is formulaic so this is you know justin fitzpatrick's opinion or something very low bond return assumptions for the

32:08

near term and then if we get to the long term um they're going up although still not you know

32:14

um spectacularly although you know you can see that standard deviations would certainly allow for some high returns um

32:20

some years so that actually when i first saw that come out um that was you know long before

32:27

interest rates went up so much i was surprised but again that's just a formulaic um thing

32:32

uh and we we have a um kind of a filter on this where we where our defaults

32:39

won't be negative um that's just a an assumption you're you're welcome to put negative returns in here um as your

32:46

capital market assumptions if you want um so i think that the formula actually spit out some negative returns for for

32:53

bonds but we they're there's a floor i guess you could say on what we'll what we'll put in as a

32:58

default right um

33:03

and then uh next question to what degree is cape ratio used in making capital

33:09

market assumptions um schiller had been saying from 2012 to 21 that cape was at

33:15

historic high in 2021 he acquiesced and said that high cape was justified in a

33:21

low rate world which has led some people to suggest that random walk may be a better model

33:26

than cape with respect to estimating future market returns so i guess can you guys speak more about you know what

33:32

degree is cape ratio used um in making cmas yeah so um it's it's basically up to you

33:39

uh how much you want to use that or not i would say again in the in in our formula for producing defaults it is it

33:46

is used um but it's not the only thing that's used so um

33:52

i can go back to that slide the

34:01

okay so so these are the seven things we're using to produce this this blue and gray

34:07

picture so um it's it's always tempting

34:13

to take economic data you know run a regression on it produce a formula and have that tell you how

34:20

much income to take um [Music] you know i think that that's a very

34:26

dangerous game to play at least you know unless you have a lot of confidence in your in your formula and

34:34

so rather than go that approach we've chosen to um to allow you to use

34:40

economic context both both for communication and for tilting your advice but doing it in kind of a

34:46

almost an agnostic um uh approach to to how this will work so

34:52

there's no formula within income lab that says if cape is high then push um income down

34:58

there's just nowhere in the code that says that what what you see here is is what really

35:05

happens if you're applying economic context which it says okay i'm going to use history but i'm not going to use the

35:10

gray stuff right and the gray stuff is just things that are more that are less like today and it

35:17

just happens that those are by and large periods where you could have had high income but there's nowhere where we say

35:22

hey exclude the high income stuff um so it's that sort of just falls out of this

35:28

concept of economic proximity um and what we're trying to do is is acknowledge that really this this is not

35:35

a perfect science you cannot there is no there's no formula that will tell you based on what cape is or what interest

35:41

rates are how much you can spend it's more like a tilt if you want to do that in your advice

35:46

you you don't have to you can you can choose not to apply this um at all

35:51

so i hope that helps and justin um derek your free answer go ahead and stay on this screen we have a

35:58

few more questions on this particular screen derek yeah we we can go ahead and jump to the

36:05

the next question um uh i didn't really have much to add beyond what justin already said perfect okay um yeah so now

36:12

uh you know regarding the economic context screen is that comparing whether any

36:17

adjustments are necessary or whether the plan fails completely even with adjustments so that is if the proposed

36:24

income line is below the great depression does that mean the client would have had to make any adjustments during that time period or that their

36:30

plan would survive as long as adjustments were made so um

36:38

it's extremely rare i mean i i can think of some some odd plans that i could make fail even with adjustments but most time

36:44

adjustments would um would save any plan so what you're seeing here is

36:50

the most because it's history we we know the outcome right so it's saying this is the most you could have spent with this

36:56

plan at this point in your life and not and never had to make an

37:02

adjustment right so this is it's trying to be a very s you know it's like a static snapshot of a point in your life

37:10

however you know the black line here you can see is slightly above kind of the early 20th

37:15

century so you would be saying okay if you spent at you know that level of just

37:22

under fifteen thousand dollars you would have had to adjust down if this were the early part of the 20th century that that

37:28

is a way to interpret this whereas um if the um

37:33

you know if those if the mountains are above that black line it's saying well in actuality you

37:39

wouldn't have had had to adjust down if that were the um period you were living through um of course you wouldn't have

37:45

known that until the end of the uh to the end of uh your plan but that's what it's telling you

37:52

and then uh next one is can you explain more about how you find similar historical scenarios um do you look at

37:58

multiple factors how much weight do you give to the different factors yeah so um we we use the seven things

38:05

you see here um and then we weight them according to their um

38:11

to basically their their importance their their power in in giving you insight into into how much

38:16

income you can spend historically so we use something called r squared um

38:22

which gives you kind of that um well i'm probably getting into trouble if i use the wrong words here the uh

38:29

i want to say predictive power explanatory power is probably the the the correct way to put it um of of each

38:36

of these um of each of these factors so you know in any period where all

38:42

seven are available you know it may be that treasury rates are you know less predictive or less explanatory than cape

38:48

and so people have a slightly higher um uh waiting in building this picture but

38:54

again the goal is just this kind of gross level picture of hey what parts of time were a

39:00

little more like today and what parts were less like today um and so in a way you don't because we're not looking for a formula that

39:07

produces a specific income based on these inputs we your your

39:12

data that approach sort of uh that works

39:18

um and the next one is how is the period that shows on the economic context screen determined is it based on the age

39:24

of the client date of retirement um can you just explain more about it so it's using the the

39:31

current plan length so if you're if um you know which is based on ages and longevity settings so if that plan

39:38

length is 25 years um then what you're seeing here is the

39:44

incomes at this point of time for a plan that was 25 years long um one thing on

39:49

the economic or the historical analysis chart that sometimes trips people up is the most recent date

39:56

you'll see here is one that was a plan length ago right

40:01

so for 25 years then it would be 25 years ago this month is the most recent one you

40:07

would see and um now to some other topics of

40:12

questions um so using the cash reserve or bucket strategy for cash flow needs

40:17

how does income lab incorporate and analyze this method for retirement cash flow needs

40:27

um let me think a little bit about that

40:33

you know there are so many bucket strategies i i guess i might have to know which what kind of approach you mean um

40:41

to think about it but by and large what's being what's being analyzed here is is kind of a total return picture

40:49

right so um it'll look at the entire portfolio if you've got a lot in cash that'll that'll

40:54

have an effect on on both the you know income proposal and the guard rails

41:01

and i can jump in and answer or at least address this from how i used a bucket strategy so

41:07

i'm i'm a big fan of and michael kids this is written about using a bucket strategy more as a

41:13

psychological tool rather than even um i mean you could use it if you're in a you

41:18

could start to spend down that um the bucket but so that that's kind of my personal approach so i'm often

41:25

talking about buckets and showing my clients buckets and if somebody does have a particularly large cash holding

41:31

or something i can reflect that in the portfolio assumption that the

41:36

information that i'm putting in about the portfolio uh but you know generally speaking it's not actually modeling to

41:42

my understanding there's no way to actually show it modeling like spending down that bucket first and then

41:48

moving on it's just going to be spending from the total portfolio is that correct justin

41:53

that's right and um

41:58

lauren if that you have any follow-up questions feel free to chat me and i can kind of promote you um

42:04

if we need to kind of get more explanation um but to the next question is uh so this is uh

42:10

to the tax center is just can you explain the break-even analysis in the tax center

42:17

sure so that's just um the point at which the cumulative

42:22

estimated taxes for um the so you have two scenario two two

42:29

approaches right a and b are one and two so that's just the point at which the taxes paid um for

42:37

the second are lower than the first so cumulatively so

42:42

um kind of more uh more practically speaking you might be looking at roth conversion

42:49

strategies because they are estimated to save you money on taxes over time

42:56

but you might wonder well how long will i have to live for that to be estimated to be

43:02

to to to to work do i ripping the band-aid off on taxes might mean high taxes in the first couple years how long will it take

43:08

me to make up for that it's it's um you know it's kind of similar um to the social security break even

43:14

right it's like well if i'm giving up something now how long is it going to take for for me to have i've decided

43:20

that was a a good idea um so it's just cumulative taxes in strategy one versus strategy

43:26

two okay um and then this next one let me

43:32

see if i can understand we may have to have them come and explain a little more um it says how would you suggest the placement

43:39

of the current guard rails um

43:44

i think a good way to to think about it is

43:52

what guardrail positioning produces both short-term guard rails

43:57

and and plan test results that are comfortable to the client um and so you

44:03

know if this is a more risk-averse client you might want that um

44:09

especially that that income decrease guard rail um to be you know farther away

44:16

from from their current position right so you know in this case that's you know that's

44:23

a pretty good drop in their portfolio maybe that's comfortable to them and then you also want to look at the plan

44:29

test and ask you know on an overall basis and kind of on a

44:36

cadence of adjustments and size of adjustments are these the sorts of things that that the client um

44:42

feels comfortable with and so you can use you know the preset the defaults which

44:48

are here just you know to to make things easy for you you know explore different settings here

44:53

there are nine preset settings or if you really want to dive in you can get into the advanced settings and and and

45:00

specify them more exactly but i'd say those are the those are the two places

45:05

um where i would look kind of the short term view and then the longer term plan test view

45:11

is is this a comfortable you know world for the client to live in derek do you have any tips there

45:17

i i uh well i think you you got it exactly right in terms of you really do want to look at that

45:22

long-term income experience make sure that aligns with the types of adjustments somebody would want to make

45:28

right if they're somebody who's going to say you know i really don't want to be making downward adjustments i'm willing

45:33

to give up some upside to do that then you're going to lean more towards the conservative side of the income

45:39

setting adjustments and you know for me too i guess there is

45:45

another factor that i'm often looking at that's also uh you know kind of a risk capacity i guess

45:52

but in an income perspective where you know some clients they given um you know if they if

45:59

they're really diligent savers they they're not spending anywhere near as high they have that ability to use a

46:04

more conservative strategy um if they would like to uh whereas other clients i may not even

46:10

you know push them so much to that option and instead frame the conversation around here's the type of adjustments that you

46:16

know historically you would need to be willing to make to make your income your desired income work and so i do kind of

46:23

filter it through that and when i run the initial plan you know i like to get at least an income target of some kind after tax

46:29

that a client's shooting for and then you know have that conversation from there if if

46:35

they're right on with kind of the middle of the road um level you know then maybe talk through that

46:42

that those types of long-term income experience and see if that's a good fit and then just go from there

46:48

if they're the target is their their personal target is way below what you know the

46:54

middle middle-of-the-road approach would say they could spend then you could bump them down to conservative and have more of that conversation about where do they feel

47:01

best but um it does lean on i think kind of those two different factors actually and this was um kind of a

47:07

clarification on the question it was um how would you suggest we explain the placement of the guardrails to a client

47:16

i personally i i tend to stay away from inviting conversation

47:23

on that topic in terms of what's really behind the scenes you know driving that of course

47:28

you know i i can have that conversation if it's an engineer or somebody that you know really is digging into the details and wants to know

47:35

um but for the most part you know i'm talking more in general terms this is a more conservative strategy this is a

47:41

more aggressive strategy that would mean bigger adjustments but more income now um and really just trying to keep the

47:47

conversation there and i think intuitively most people grasp that that you know they could take more income now with more risk of

47:53

adjustment and that's where i'm framing most of my conversation thank you okay um and just justin while

48:00

you're on this page this one is real quick um where can you see what cmas are being used in your household analysis

48:08

so that's a great question if you go to advanced settings

48:14

plan analysis you've got this historical if you go to default

48:20

values um you can set what your default is so any new plan any

48:26

new household this will be the setting you can always go and change that um so if you say well you know i've got my

48:32

regime based money carlos cmas i've got my process for that that's what i want to use you could set

48:37

it here and then any new plan you create we'll use that keep in mind anything you've already created we we won't

48:44

overwrite that you know generally we don't want to make changes behind the scenes that you're not aware of so um so this is the

48:51

default values are only for you know future things you create if you want to change them on existing plans

48:57

you want to go to the advanced settings and then this next one um

49:03

i'll throw it out there but i may kind of say it may be best for schedule a meeting um directly to kind of talk

49:09

through this scenario but the question is help me understand how an income of

49:15

and fifty seven dollars per month um only has a tax rate of six percent um

49:20

the f meaning the after tax income is thirteen two eighty one um

49:26

so i think the question is is uh is our cash flow enough

49:32

to pay all the taxes or i think maybe helping them understand kind of

49:37

what maybe why the tax rate is potentially so low with their before and after tax amount

49:44

yeah i mean i'd have to to see all the cash flows that are in the plan to know for sure um

49:50

but you know certainly there's a certain amount of uh income

49:55

that is tax-free for for anybody right even if it's just just because of your your standard deduction um but there may

50:02

be other you know return of return of uh of basis um

50:07

could be um you know roth distributions all sorts of things and that's in the denominator

50:12

right so we're doing total taxes over total um total income um and so those could be

50:19

could be parts of it i think uh one point or just quick clarification i think they're actually talking about this scenario up on that

50:26

on the screen so explaining that the before and after text is here i'm glad you're putting that out derek my uh

50:33

q a was never doing that we can we can dive in um

50:40

take a quick look um

50:46

[Music] yes it looks like the way that this is

50:51

put together there's a bunch of um non-taxable income this is just a random

50:57

example household so i do not recall what i put in here but you can see you know even the taxable social security

51:03

takes a while to get above um uh into the 10 bracket so i'm sure

51:08

that's what's what's going on there okay um we've got about six more so i'll just

51:14

kind of keep going through these um is there a tutorial or video of how kind

51:20

of this would be presented to a client um i could potentially answer that our next webinar will kind of just talk more

51:26

about how to have this conversation and kind of present income lab to clients um so i

51:32

think for keith we don't have a specific tutorial video but i think our next webinar um next week tuesday would

51:37

actually kind of help answer and give me more um background there on how um kind of

51:44

derek more specifically is even having this conversation um with clients

51:49

was that a is that a fair answer do you have anything else to add perfect i will

51:55

consider that one done okay uh so from there you know um so kind of back to more modeling how would you suggest

52:02

modeling portfolios that use alternative assets like gold manage futures private equity um etc so imagine that's 20 to 30

52:10

percent of the total portfolio um inside the software um and then you know how would you model replacing that portion

52:16

of the portfolio um oh with what how much oh i guess if um if we don't have that

52:22

you know kind of what would portion of the portfolio would you kind of replace um to factor those in

52:30

yeah i mean this is always um you know i guess it's not technically

52:36

correct usage but i always think of this as like basis risk almost like right you have to choose something

52:42

to in this case model the world with and then you have an actual investment right so sort of like if you're doing

52:48

hedging you have you know your your derivatives and your actual thing like what's the connection are they going to move in the same direction are

52:53

they going to behave in the same way so um it's you know choosing from the asset

52:59

classes available which ones you think are you know in in the case i guess you know

53:05

best you know at least bad um as a as a as a way to to model that

53:11

i think the expansion of asset classes is going to help that some but there are asset classes always that

53:16

won't have a an exact mix there derek have you that i'm sure that's a

53:23

common issue not just with income lab but lots of software programs have you heard

53:29

other uh takes on it i mean to me that that's about as good as you can do right to just trying to

53:35

find the best alternative and realize it's not perfect but also you know at the end of the day no

53:41

no none of this is perfect we know that it's not you know it's a tool and understanding the limitations of the tool and how you

53:47

use it with your client i wouldn't personally be afraid of using a tool just because it doesn't perfectly

53:52

match the asset classes i'd just be cognizant of that and realize that as your guard rails get updated at the

53:59

end of the day that's still going to be the most important thing in terms of how to adjust that spending

54:07

okay next question does the program assume a constant asset allocation throughout pre-retirement and post

54:12

retirement or is there a way to account for the fact that you know as they approach retirement the clients may be

54:17

reducing their equity allocation that's a great question it is a it is a

54:23

constant um asset allocation but that is a a feature that many people have asked for and so it is

54:29

something that we're that we're looking at adding um and so really the main question is

54:35

for us add in adding that feature you know how best to cover as many possible use cases as as possible right um

54:43

because there may be lots of ways that people would want to to model that whether it's glide paths or changes at a

54:49

particular point in time and so on and then um how do reverse mortgages

54:55

life insurance loans and inheritances fit into income lab

55:00

outputs yeah so mostly you're going to want to put those in as cash flows

55:07

um under other income um so you know i could i could put in an

55:16

inheritance i guess is a little tricky because you have to you do have to put a date um oh and i even misspelled that

55:22

um you know so i'd do that one time be a little bit conservative maybe on the

55:28

date um and you know we could call that non-taxable

55:34

right [Music] um same sort of thing with with reverse

55:39

mortgages but there you'd you'd be looking at a recurring um cash flow

55:46

and um is a fire client so i'm assuming it's uh financial independence retire

55:51

early client able to be models in income sure absolutely yep just uh

55:58

you know using the their ages and or their dates of birth and the longevity settings will adapt to that

56:06

and um and they'll have a very long plan

56:11

and this last one is not really a question but just feedback and it's one of our users said uh my big attraction

56:17

to income lab is that it seems to be the first and best tool for implementing dynamic distributions and thank you

56:26

always love the feedback so yeah we really appreciate that um and then any last questions um we have a

56:34

few more minutes here so we probably have time for one more i will see if something comes into the

56:40

queue okay

56:45

well justin derek again guys thank you so much excited for this next year of webinars um and and amazing topics to

56:53

come um for our users thanks for joining i will send out the recording um

56:58

so uh folks can view other um later and we will um have our another

57:04

webinar um next tuesday and we'll kind of let folks know about our upcoming webinars in february march outside of

57:09

that um for anyone who has kind of specific um scenarios or questions that you want to help with please reach out

57:16

to your account manager or to our team um and we're more than happy to hop on a zoom and kind of review any um specific

57:21

scenarios or just help out with any questions that come up as well and with that guys thanks again and i

57:28

will go ahead and close the webinar have a good day everyone bye

57:34

bye