Do financial plans underestimate inflation risk? How to handle sequence-of-inflation effects in planning?

um so let's get to today's topic

0:06

um which is um which is on inflation and financial planning especially retirement

0:13

planning um and this if you're interested in more this material all

0:18

comes from um an article that I did on the the kitus blog kits.com nerds I view

0:27

um I think maybe a month ago or so um so we can include that link when we send

0:32

the followup to this um to this webinar if you want to kind of read read through

0:39

some of these materials I've added a few things to these slides that aren't there and we of course happy to take um

0:45

questions at the end about this um so this um this topic is really about how

0:52

to think about and uh inflation in retirement planning and um and kind of

0:59

get at what some of the some of the problems are with the way that we currently treat

1:05

inflation um in in uh retirement

1:11

software um so I I think it's really useful to think about inflation uh risk in retirement and

1:19

compare it to Market risk because we're really used to thinking about Market risk in in um in retirement planning

1:26

it's something that although you know you could certainly quibble with uh Little Bits here and there on how we do

1:32

the analysis and so on I think it's a place where we do reasonably well as as uh financial advisers handling Market

1:39

risk and its effect on um on retirees on what they can spend on their possible

1:45

experiences and so on so the key things to know about Market risk are returns

1:50

themselves if you are invested as a as a retiree if you have some money that's invested in stocks bonds and so on those

1:57

returns are variable so we don't expect them to be exact the same every day every month every year and they're

2:02

uncertain so we we don't know ahead of time exactly what they'll be although we might have some reasonably good

2:09

estimates and guesses at the range that they might have um that means that

2:15

return sequences can be better or worse than expected um typically people think of this as a

2:22

return sequence for a retiree is better if the good returns are early or if

2:28

early on somebody has good returns um of course it's always better to always have good returns but if you're going to have

2:34

good and bad returns it's better to have the good ones early um we also know that failing to account for Market risk could

2:41

really lead to a a bad financial plan um if somebody is invested so specifically

2:47

failing to account for the variability and the uncertainty would typically lead us to underestimate risk or overestimate

2:55

the spending that someone could have and you know like I said I think as financial advisers um both our

3:02

technology and our you know our knowledge and thinking about this are pretty good uh we we've we've kind of handled this um this part of uh of the

3:09

risks to retirees another important risk is not these are not the only two risks in retirement by the way but another

3:15

important one is inflation risk and inflation itself is very much like uh

3:23

returns in in these senses it is variable and uncertain so we should not expect inflation to be identical day

3:30

after day month after month year after year it's uncertain in the sense we can't perfectly predict it um in fact

3:36

it's not all that easy to predict at all but we probably know you have reasonably good estimates of the range of of

3:41

inflation experiences we might see also inflation sequences can be better or

3:47

worse um for retirees uh and and in this case

3:53

again having good inflation which this means lower inflation right whereas for returns good returns are higher um good

4:00

inflation early is better um of course it's better to always have low inflation

4:06

for uh for retirees um but if you're going to have bad inflation meaning High inflation it's better to have it toward

4:11

the end um and again failing to account for inflation risk can lead to underestimating risk and overestimating

4:18

spending capacity um we'll kind of have this discussion

4:24

around should we be treating inflation in um

4:30

in our financial plans as fixed or variable and in a way again comparing it

4:36

to investment returns Market risk um the the world of advisor technology and

4:42

planning technology has has kind of always gone from um fixed just just you

4:49

know treating returns as um steady and

4:54

and at one level throughout the the plan that was originally an in certain

5:00

situations still is how we might um we might do projections in retirement and

5:06

this is really just using you know financial calculator type math um maybe it's you know Excel time value of money

5:14

functions um right saying okay we've got 30 years or 360 months and we expect

5:20

this level of return you know what can I spend um and that again there are places

5:27

where doing that makes sense but that's typically for kind of practical reasons

5:33

when you have to show someone you know particular numbers over time if you're trying to understand risk and really

5:40

give people good advice on what they can spend in retirement that that takes into account that risk treating investment

5:47

returns as variable and uncertain is really the way to go um and and

5:53

so what I'm going to argue in this in this webinar is that that's also true

5:58

for inflation so we should really be treating inflation as as variable not as as fixed um so typically in planning

6:06

technology today um inflation is fixed which means it has the it's the same uh

6:14

it's the same amount every year there are technologies that have different types of inflation maybe you have you

6:20

know core inflation and inflation for healthare and maybe inflation for uh education right different kinds of

6:27

inflation but that those are all still fixed so they're not varying year by year or month by month um and one way to

6:34

know this by the way is just just look in your technology and see whether there is a standard deviation for inflation or

6:40

a variance um if there's not then it's fixed um the thing is though we uh well

6:48

we made this transition um over 20 years ago in financial planning from you know

6:55

treating returns as fixed and known to treating them as variable and uncertain uh we are are yet to make that shift in

7:03

the world of inflation um so of course that income lab we have made the shift

7:09

but I I wouldn't say that the shift you know people have fully embraced um treating inflation as fixed uh as uh as

7:18

variable and uncertain so the the place to start is you know is inflation variable and

7:24

uncertain um I think it's it's probably fairly obvious to everyone but just to make it clear here um yes inflation is

7:31

variable and uncertain here's the last 100 years of inflation uh annualized inflation rates um and you can see

7:39

there's been quite a range um even in just the last um the last I don't know

7:47

decade or so maybe a little bit more than that um you can see the most recent peak of inflation was in June of 2022 uh

7:55

the most recent deflation was in April of 2015 so only a 7 year gap between

8:00

those and we saw more significant deflation in July of 2009 so this is all

8:06

within you know recent living memory we've seen quite a range of inflation so

8:12

this isn't the kind of thing where we look back at you know at history and we say yeah okay fine in the 40s and and

8:18

30s and before that yeah we saw some crazy inflation but you know this that's not the thing the sort of thing that

8:25

happens anymore I think um it it's easy to remember from kind of 000 to 200 um

8:31

you know 20 or so it it did feel a bit like that like hey maybe maybe times of

8:37

of crazy changes in inflation are over um I think 2022 showed no that's that's

8:42

just not true um so just just for to give you a

8:48

feel for what inflation has been like over this last 100 years um it's averaged about 3% a year with a standard

8:54

deviation of about 3.9 um of course different periods would have different um averages and standard

9:04

deviations um so uh that shows that inflation really is uh variable and

9:10

uncertain but the but the next question is okay but does that matter for

9:16

retirement income planning um because let's face it anytime we do planning uh

9:21

it depends on kind of a model of the world and certain decisions about you know how we're going to model things and

9:27

so on and that will always include some kind of simplification

9:32

right so maybe this is just a simplification that's fine um now I I think since we we know that that

9:39

simplification is not okay in the world of um of investment returns of the the

9:45

returns that you get in your portfolio um I suspect everyone will probably s uh

9:50

will will guess that it's also important here but you know it's it's worthwhile

9:56

checking checking to see what affect treating inflation as fixed and known

10:01

has on Financial Planning and does it have a negative effect so one way to to

10:08

try to answer that question is to find a situation where inflation is really

10:14

important uh and ideally it's it's one of the only things that's important so we're kind of you know taking this into

10:20

the lab and saying all right can we can we build an experiment where um where we'll be able to figure out how bad it

10:27

is to treat inflation as fixed and known um so here we have a um a

10:32

situation you're unlikely to experience out in the wild um but it's it's a family who is funding their retirement

10:40

um only from a pension that's not adjusted for inflation um this is a $10,000 a month pension um nominal no

10:48

adjustments for inflation let's assume it's joint life as well so that we don't even have mortality risk here um and if

10:56

we assume that its inflation is going to be 3% we assume that they need 30 years

11:02

of of um inflation adjusted income we can do the math relatively simple on

11:09

simply on how much of that $10,000 they can afford to spend to begin with

11:16

adjusting it for inflation over time um and it ends up being um

11:21

$664 a month at the very beginning of retirement um so they spend

11:27

6,640 3,360 goes into um a savings account or

11:33

a CD so let's let's assume also that this family is very risk averse and it's

11:38

just investing in um you know interest bearing uh instruments that exactly

11:44

match inflation um so we're going to imagine that we can we can just always get an inflation rate that that just

11:49

keeps us up with inflation and that's it um so that's uh you know that's a

11:55

relatively maybe shocking uh number for people right they may see that and say hey wait a second I literally get

12:01

$10,000 into my bank account every month and you're telling me that a third of that has to uh has to be put aside for

12:08

the future um and the reason is over time they're going to uh slowly adjust

12:14

so that in nominal terms they won't be spending 6640 every every month or every

12:19

year uh they're going to increase the amount that they spend decrease the amount that they save um in order to

12:26

keep up with inflation and then in year 14 here um they'll flip it around and start

12:31

spending the full $10,000 from their pension uh and drawing from savings to make up for uh inflation

12:39

that's happened uh over over time so if we knew inflation was going to be 3%

12:45

steady uh and we knew that it was going to be a 30-year period this is exactly

12:50

what we would do everything would work out great um and and there you

12:55

go the problem of course is that's not actually how inflation behaves so for this kind of you know uh specialized

13:04

situation we can look back at um how inflation has behaved historically the

13:11

actual sequences of inflation and ask given those inflation experiences some

13:17

of them which were really high inflation some of them which involved deflation um some high inflation early some high

13:23

inflation late and so on um what's the range of real spending levels from this

13:29

$10,000 a month pension that people could have sustained historically and we

13:35

can rank those you know according to um according to their their um the the

13:43

percentile of history that would have survived this level of spending so think of this as like probability of success

13:48

but using historical um inflation sequences so 50% of the time you could have spent

13:57

um almost $6,400 a month um if you

14:03

wanted to have a spending level that would have survived every single inflation sequence in history you would

14:10

have had to spend $4,350 so save more than half of your

14:17

pension in order to make up for future um inflation adjustments or to provide

14:23

for future inflation adjustments in the future um this lowest a number here

14:29

comes from April of 1973 so you know thinking back there that makes a lot of sense um because we

14:37

were really just entering the the teeth of stagflation in the 70s and 80s and um

14:44

you know really uh we would have had to put a lot of dry powder aside to uh to to make up for that kind of inflation

14:52

sequence over time um so what this shows us is if we

14:58

simply assume a 3% fixed inflation which by the way is about average for the last 100 years um we end up with a spending

15:07

level recall it was 6640 um that is actually

15:13

higher than um even the 50th percentile historically in other words historically it had less than a 50% success rate um

15:22

and so that should be concerning to us it shows us that you know whatever the

15:27

range of infl can be um it it needs to be taken into

15:34

account or else we would be you know recommending to this family a spending

15:40

level that that seems to be much much higher than we would otherwise have felt

15:45

comfortable recommending so if you think about if if this were a um a case where

15:50

someone was withdrawing from an Investment Portfolio we would have looked at the

15:56

range of possible sequences of returns that they could have right good bad medium different times and said okay

16:04

let's find ourselves us a withdrawal level that you know is is reasonably

16:11

conservative if they can afford it right so typically people might look for 70% 80% 90% probability of success that kind

16:18

of thing um because people are risk averse and we'd rather you know be surprised on the upside than on the

16:24

downside well if we took that into this situation we would be recommending you

16:29

know things around $5,000 a month so saving half of the uh of the nominal

16:36

pension um if we took you know historical inflation experiences

16:44

seriously um now you might say okay well maybe that 3% inflation that was just

16:50

too low right I mean maybe it is the historical average but we know that if you want to handle variability

16:59

um we typically need to not just use kind of a arithmetic average in things

17:05

but we need to maybe maybe just bump it up a little bit so we can look here and ask okay what fixed inflation assumption

17:13

would I need to use if I wanted to um give this family advice to spend say

17:19

around $5,000 a month from their $10,000 pension and we can see here um that it

17:24

is uh you know we would need to get into the five and six% annual inflation range

17:30

um if you wanted to get up to 100% closer to seven um% inflation um so

17:36

there are a few problems with this the the the main one is really for communication with clients this this

17:42

seems really high right now maybe you do expect six or 7% inflation for the next

17:47

30 Years but uh there aren't many people who do and so this would be hard to explain to clients um it's also not

17:55

actually the way that inflation works um instead inflation is variable and

18:01

uncertain and so um what we're doing here by just by just increasing the

18:07

inflation assumption is we're still only able to produce one answer to the

18:14

question what can I spend from that $10,000 a month um uh uh pension so unlike in the case

18:23

of withdrawals from a portfolio where we kind of say okay yeah I could withdraw a range of things just the more I withdraw

18:28

the higher risk I have you know let's find a kind of a risk return tradeoff that that works for these folks in this

18:35

case if you plug in you know 6.2% inflation that's great but you just get one answer you're not getting a a range

18:42

of of experiences um so to compare this to if we if you

18:50

wanted to do the same thing um for withdrawals so not use Monte Carlo or

18:56

historical analysis or use you know varying returns you could um so for

19:02

example if you if if you normally would have assumed an average real return of 5% with a standard deviation of 10 um

19:10

and you had a million dollar portfolio and you wanted an 80% probability of success um you would have recommended to

19:16

someone that they spend uh that they withdraw $46,200 a

19:22

year you could get that same answer just using a financial calculator but you

19:28

would have had to assume a 2.3% fixed real return using time value of money calculations right so it is possible to

19:34

do this as well in the in the world of of portfolio withdrawals but it's not what we tend to do and the main reason

19:40

is it's not really the way the world works right instead you actually get um varying returns it would be kind of hard

19:47

to explain to clients why it's 2.3% even though I actually think it's going to be average of five right this would just be

19:53

a really convoluted you know statistical conversation um and it just produces one

19:59

number it's this 46,700 really what I want is to understand that yeah there's a range of behaviors I could follow a

20:05

range of plans I could follow they just have different kinds of risks so for

20:10

making decisions about how you know what to do in retirement how much you can spend that's where we want to get we

20:16

want to get to a place where we where we get a range of possible things we could do and then just make informed decisions

20:22

about that risk uh tradeoff so I think this idea of just using higher

20:28

inflation assumptions um is not is not really a good way to go um because it

20:34

has all the same problems that using a really low um return assumption would have in uh in withdrawal

20:43

analysis um sort of an aside on what inflation risk is really like and where

20:50

it comes from here um we often talk about inflation risk and we just sort of

20:56

think it has to do with well what what average inflation will you experience um over some period of time um and

21:04

certainly periods that are worse for retirees for with with inflation do tend

21:10

to have higher average inflation but it's that's not actually exactly what's

21:15

going on just like with returns what we're actually experiencing here is sequence of inflation risk so um what

21:25

that means is it what's what really matters to to people's experience is

21:30

what is inflation and at which point in in their life again if you have high

21:36

inflation early on that is can be a problem High inflation later is not as

21:41

big of a problem and so what you can end up seeing is for example um you know

21:47

this uh this uh let's see here the 70th percentile here the 5800 a

21:56

month um that that's lower than the 60th percentile right which makes sense so

22:02

it's a higher probability of success uh 60% is almost $6,200 so about $400 more

22:09

but if I look over at what the actual average inflation was during this period

22:14

the average inflation for the $5,800 is lower right that might look like a

22:20

mistake but but no it's it's definitely possible to have high enough inflation early on that the sequence of infl is

22:28

bad but then maybe inflation later later on is is low enough that if we average the whole

22:33

thing well it's it's 3.4% instead of 3.7 um so this is why it's if we're being

22:40

precise we probably want to talk about sequence of inflation risk just like we talk about sequence of returns risk that's what it's all about um another

22:48

case is right here um we see you know the 80th and 90th percentile again about

22:54

a $400 difference in spending capacity but they both have the same average

23:00

inflation over that 30-year period so looking at just average inflation is not

23:06

what's going on here what what really is is sequence of inflation um it's this is really nice because it's really the same

23:13

as if I we're doing uh you know sequence of return risk the only difference is uh

23:18

it's it's flipped on its head so it's we want low rates of inflation early on that's good sequence of inflation uh

23:25

High inflation early is is bad SE sequence of inflation okay so how should we handle

23:34

sequence of inflation risk when we're doing planning um I mean it should be fairly

23:39

obvious just like we did with um with returns when we shift from you know flat

23:47

fixed returns to variable returns is we need to treat inflation as variable and

23:53

uncertain that in practice typically would mean either using money car

23:58

simulation where inflation is not the same in every month or year of the

24:03

analysis um where we have it an average inflation and a standard deviation of inflation or using historical simulation

24:11

um as I did in in the examples earlier um to to use actual sequences of

24:17

inflation over time um this will allow us to produce spending advice retirement

24:24

spending advice um from a range of possibilities where we can really say Okay are our

24:30

assumptions good and if they are good you know how much risk do I want to accept versus protect myself against so

24:36

for example here I did a a 30-year um real spending analysis using Monte

24:42

Carlo um where I assumed uh I believe it was 3% average inflation with 3.5%

24:49

standard deviation same $10,000 month pension not adjusted for inflation and here we see a

24:56

range from you know 100% success to 50% success on what that analysis says this

25:03

family could spend from their $10,000 a month pension um so at this point we we

25:09

would look at okay if if if we're really worried about having some of the worst inflation sequences ever we may be

25:16

spending less than half of our pension to begin with here on the other hand maybe you're saying well you know I I

25:22

don't think that's quite as likely right so maybe we to go more toward the 80 70% um may even be that you think um you

25:30

know if I if I threw the retirement smile in here for example that maybe they actually can absorb some inflation

25:35

over time because their spending will go down as they get older okay well then I might be able to allow them to spend even more or recommend that they can

25:42

spend more but now I'm I'm back in the world that that we're familiar with from giving retirement advice which is the

25:48

world of trade-offs the world of figuring out what risks are you know how important they think that we think they

25:53

are and also you know clients abilities to um to adapt to adjust um if things

26:00

turn out worse than expected so we're not at this place where we're sort of pretending we know exactly what's going

26:05

to happen and just giving uh you know basic calculator type uh

26:12

advice okay um so I'm going to now go to a new

26:18

example but just to um just to reiterate how to tell um you know whether your

26:25

software is treating inflation as um as variable and uncertain or as fixed and

26:33

um and known um just look to see whether you have um access either to historical

26:41

simulations that use actual historical um inflation um or whether you have

26:46

Monte Carlo and a a standard deviation or a variance assumption um in the

26:55

software if you don't then it's being treated as fixed and known even if you

27:00

have multiple uh inflation assumptions that those are just you know different inflation for different categories of

27:06

spending but they're not varying inflation they're not treating it as a risk um and maybe that's the other point

27:12

to make here is this might sound a little bit more um you know controversial but if if we're not

27:20

treating something as variable and uncertain we're not treating it as a risk so when I treat when I say

27:27

inflation is going to be 3% um that is not acknowledging that there

27:33

is inflation risk that's just if I know what the inflation is going to be just like if I knew what the returns are

27:39

going to be I can tell you exactly what to do there's no risk involved risk is involves uncertainty things that are

27:46

unknown um so I think you know in a technical sense by not having variation

27:52

in in inflation in retirement analyses we are saying that inflation is not a

27:58

risk which is clearly false okay so that example that we had

28:05

before of the $10,000 a month pension you know as I said that's that's not common to run into someone who has only

28:11

a $10,000 a month pension uh I have certainly seen cases where people have significant um not adjusted for

28:19

inflation pensions as part of their plan I've seen in particular remember uh looking at several or talking with

28:25

advisers about several cases in in California where you know maybe it was a kelper pension or something I'm not an

28:31

expert on those pensions but uh significant you know $6 to $10,000 a month in not adjusted for inflation um

28:38

pension so these these kinds of scenarios certainly show up but on the other side we might say

28:45

well you know my typical client is just withdrawing from a portfolio and they have social security which is adjusted for inflation so maybe this is not that

28:52

big of a deal for me um and at the end of this um webinar we'll definitely go

28:59

over okay what are the sorts of things that make this make it more important to

29:04

treat inflation as variable and unknown um but we can first go to the just the totally other end and say okay what

29:10

about a plan that is only based on portfol withdrawals again this is not particularly common to find either

29:17

typically people at least have social security um but it's a it's a good

29:22

question because this is a place where we already deal with Market risk and we

29:30

already have standard deviations built into our Capital Market assumptions and so you might wonder

29:36

well is that already kind of handling the variability of inflation um and and the answer is um it

29:47

depends so in order to do this study we we had to look at we used historical in

29:55

um returns again and historical inflation because this captures you know exactly how inflation and returns

30:01

actually interact with each other um but we did one where we took the nominal returns but but average inflation and

30:08

another one where we took nominal returns but actual inflation and you can see that across the board using actual

30:16

variable inflation produced lower withdrawal rates on a on a portfolio I

30:23

believe this was a 50-50 portfolio stocks and bonds um so it does look like

30:30

um there is still an effect of using fixed versus variable inflation and the

30:37

effect is a tendency to understate risk and overstate possible withdrawals

30:43

possible spending if we're treating inflation as fixed however as you'll see here uh some

30:50

of these differences are not that big right in particular at 50 50% success they are it's basically the same

30:57

um so let's look a little bit uh deeper here now I know this is a really complex

31:03

chart I'm going to pull something out here uh shortly just to look at but the effects of using variable inflation

31:11

versus fixed inflation depend on um the returns that you're projecting or

31:18

another way to think of it is at it depends on the stock allocation or Bond allocation and it depends on you know

31:26

how aggressive or conservative you are being um with the advice so are you are

31:32

you targeting you know really high probability of of success like 90% success or are you you know more toward

31:38

the 70 60 range um I'm going to pull out just one example here we'll go back to that chart

31:45

but for example if I look at um stock

31:50

allocations between 25% and 75% so the rest of it being bonds um and I look at

31:56

90% success um using fixed inflation these are the annual

32:02

withdrawals I would I would get so from a million doll portfolio 25% stock

32:07

allocation 90% success I would be recommending $ 42,300 in withdraws um same exact

32:15

success level but 75% stocks I'd be at over $50,000 right so this is we're

32:21

holding everything constant except stock and bond allocation as I increase the stock allocation I get to spend more at

32:28

the same um at the same probability of success if I then ask okay yeah but what

32:35

if I treat inflation as variable what's the probability of success now of each of these withdrawal levels each of these

32:41

spending levels um that 42,000 now has a 72% probability of

32:46

success so 90 versus 72 when I have a higher stock allocation

32:52

higher returns that 90 goes to 80 so all these I think are significant but but you can see they're they're bigger if I

32:59

have more bonds lower return expectations um and so on so back to

33:05

this chart in fact if I'm 100% stock 70% probability of success there's

33:12

no effect in that case of using variable inflation I get the same result um so

33:17

the the spending level for uh fixed inflation analysis uh has a at 70% probability of

33:26

success also has 70% probability successes by a used variable so you can see there's like there's a different

33:33

amount of bang for your buck um if depend for using variable inflation

33:38

depending on exactly what you're doing with the

33:43

plan and in general this is how I would summarize these

33:49

effects um higher Bond allocation lower expected

33:54

returns means more and more important that you treat inflation as

34:01

variable lower risk spending levels or higher probability of success so if

34:07

you're trying if you feel most comfortable having people spend at a relatively low level um in order to kind

34:13

of build themselves a nice risk buffer you know decrease the chances that they'll have to adjust their spending

34:18

down mid retirement and and so on um that means you have a there's there's

34:25

more value in having variable inflation

34:31

um if you have more not adjusted per inflation portfolio income so this example of a of a nominal pension uh

34:38

that makes it more important that you have variable inflation in your analysis and also if you expect higher inflation

34:45

or higher variance of inflation it's more important so obviously if I really think inflation is going to be 1% with a

34:51

you know 0.5% variance you know this is probably not that big of a deal um

34:57

why is this why why are these the um the patterns that we

35:02

see um for the first why are lower expected returns higher Bond allocations

35:09

um more sensitive to varying inflation well

35:16

it's because um when you have you know say your bond allocation uh maybe you're

35:22

assuming you're going to get you know three or 4 perc returns well if you happen to have some periods with high

35:28

inflation that eats that return away very quickly as opposed to you know an eight or nine% return for a stocks or

35:35

even 12 13 um it just has a bigger effect it's taking away most of the

35:40

return as opposed to if you're at 12% stock return you know even a 3 four five it's just not as big of a proportion of

35:46

your return that's that's being taken away um and this is what we see um historically is that you know the bond

35:53

returns tend to suffer more in uh in these periods of high

35:58

inflation why would it matter that you're targeting a a lower risk spending

36:04

level or a higher probability of success why is 90% success more affected by

36:09

variable inflation than 70 um here it's because um if if you think about kind of

36:17

the scenarios that are being created by Amonte Carlo or by historical um

36:22

analysis the actual sequences of returns that correspond to these you know these

36:28

high probabilities of of success they're worse right they're worse sequences of

36:34

inflation just like if you're looking at a 90% probability of success for for

36:39

withdrawals those are worse sequences of inflation if you look at the actual sequences those are going to be um

36:47

places where you had high inflation early um and so um whereas when I get

36:53

closer and closer to you know 50% probability of success that is by definition that's average that's the

36:58

median um and so we should expect in you know the sequences of inflation that

37:03

that are are actually appearing in those scenarios to be more average more run-of-the-mill um and so we wouldn't

37:11

expect them to have you know such a such a um a difference when I'm doing variable inflation versus fixed because

37:18

you know at a 50 percentile roughly we should actually expect variable and inflation variable inflation and fixed

37:23

inflation to to be about the same we wouldn't really think that there'd be any any major difference you you'll see

37:28

some differences but that's just like noise it wouldn't be um you know something that that you've really

37:34

discovered and so by it is the tendency of most advisers and I think this is very reasonable to Target lower risk

37:42

spending levels if somebody can afford it um because now you're planning for Worse um sequence of inflation and

37:49

sequence of returns and then if things turn out better than that that's great you have you have good news for them as opposed for planning for you know

37:55

something at kind of the median of the average and now uh oh could go either way and I might have bad news for you in

38:01

the future um so unfortunately what we what we found here is um you know if

38:07

people do have bonds and in any sort of you know reasonable amount um you know

38:13

even 25% um if people are targeting lower risk spending levels using variable

38:19

inflation is more important and that is actually where most people are so um although there are some scenarios where

38:25

it's less important um unfortunately most real world situations um it's probably a good idea to use variable

38:33

inflation um so people have asked us you know because of this research okay well what do you do in in income lab um you

38:42

know we offer different ways of doing uh the the analysis so we offer historical

38:47

um analysis which uses actual sequences of inflation um we offer traditional and

38:55

regime based money Carlo both both of which have standard deviations for inflation meaning when we are analyzing

39:02

you know the the the real um you know today's dollar spending capacity of a

39:08

nominal pension it's you know some scenarios it it buys you a lot some scenarios it quickly it's purchasing

39:14

power is quickly quickly um eroded right and then we're able to kind of say all

39:19

right where do you want to be on this um risk versus spending tradeoff um and so in particular we've

39:27

seen this quite a bit where people like I mentioned maybe they have $6 to $10,000 in um nominal pension income

39:36

we've a lot of people have noticed wow this you know especially as I go toward

39:41

higher um uh or you know lower risk um you know more more secure kinds of

39:48

spending levels um it seems like I'm really having to tap back my my spending

39:54

from a pension like that um that's because because you we are doing sequence of inflation analysis and it's

40:00

saying okay if if you really want to protect yourself from inflation risk sequence of inflation risk you're going

40:05

to have to spend a lot less um whereas if you want to accept more of it you can you can go toward a higher a higher

40:11

spending level um so I think with that we can go to um to some

40:18

questions Taylor if you've had a chance to look at those yeah we have a couple that are upvoted um we have quite a few

40:25

questions so um if you want if you can I'll just go off the upvoted ones so do

40:31

you think the government overspending and the increased budget deficits will change taxes and inflation levels for

40:37

retirees now um I I don't know that I have a an a a

40:48

great you know crystal ball on that kind of thing I mean I think in in the past

40:53

uh we we've certainly seen uh there was even in a period where I think people thought that you know higher government

40:58

spending wasn't going to lead to higher inflation clearly that was shown to be wrong over the last year so um I I'm

41:04

probably fairly uh um you know I'm not I'm not the kind of

41:11

person who likes to make predictions especially about the future um so that's uh yeah I wouldn't be surprised if we

41:17

saw taxes go up or or you know inflation vary I kind of when I look at the last

41:22

hundred years and I see spikes and valleys and um you know I I I would not

41:29

be shocked if we continue to see that kind of thing over the next many decades and as a reminder if you see

41:36

questions in there you want to hear the answers for upvote them as we might not have time for them all so the next one is given the given the impact from the

41:43

variability of inflation on safe spending amounts historically varies with the allocation of the portfolio

41:48

does income lab have or plan to have a tool that lets users easily evaluate how

41:54

portfolio allocation impacts spending capacity I.E bang for the

42:00

buck um so I definitely think you can you can do that already in a sense where

42:06

you can do different allocations if I'm understanding the uh the question right

42:11

you can do different allocation so I've definitely seen people you know create two or three um plans that are identical

42:18

other than the portfolio allocation so keeping inflation assumptions the same and so on and then you will definitely

42:24

see the difference in you'll you'll typically be able to spend more at least initially if you have a higher stock

42:31

allocation but if you go into the retirement stress test um and in particular if you hit it with

42:38

stagflation every time I've seen this if you go from kind of 6040 to you know 8020 or something like that you increase

42:45

the volatility of the income itself of of what you can spend so in a way what you're saying by giving someone more Equity is

42:53

saying hey are are you the kind of person who wants to to be able to spend more when things are good but is fine

42:58

spending less when things are bad right that's that's this trade-off whereas a more moderate portfolio might be hey I I

43:04

get it I I'll spend less today but I'll probably have less volatility when things are bad that's that's the

43:10

trade-off and you can already do that and I think the retirement stress test is a great place to do

43:15

that the next one is should we use CPI or CPE inflation for

43:21

retirees yeah I mean I know there's been different and even the government has used different ones over time um in

43:27

order to get a long history we used um CPI in this in this work and that's what

43:33

we used to create our um default Capital Market assumptions you're welcome to to use different um indices uh if if you

43:41

think one works better for retirees for sure and then the next one that has the

43:48

most up votes is since inflation and market returns are not independent how would varying both in analysis and

43:54

simulation work yeah so um the nice thing about using historical

44:02

analysis is that the the non-independence the dependence of uh of

44:08

inflation and returns is just built in like nobody has to uh create all sorts of crazy um Capital Market assumptions

44:15

involving second and third order interactions of inflation and different asset classes um so that is one reason I

44:22

used historical in this research um it certainly would be possible to do the

44:28

same thing with Monte Carlo but there's a there's a point at which it's kind of um it's not very practical to ask um

44:37

advisers and firms to produce all those different um you know second order correlations and autocorrelations and

44:44

things um so we uh you know we in our

44:49

Monte Carlo we just treat them as as independent um it's you know as you say

44:54

probably not exactly correct but but um it gets you you know it gets you somewhere and it's a a nice a nice

45:01

balance between practicality and uh and at least getting you some of the effect of of uh varying inflation over

45:09

time so I think what what was that the the implication there was hey you know

45:14

we would typically see you know low real returns when inflation is really high which is true that's that's exactly what

45:21

happened so there tends to be a negative correlation between um

45:26

asset returns at inflation uh real returns anyway nominal returns there's basically no

45:32

correlation and in the same bucket if a client is invested in stocks won't the increase in investment value offset the

45:40

increase in inflation yeah I think that's exactly what I maybe I didn't say as clearly as that I like that but

45:46

that's basically what is going on um here so you'll notice over on the right that the the these are these are in

45:53

groups so you have uh blue red groups all right and way over on the right 100%

45:58

stock you see the gap between blue and red is is much smaller uh than if you go all the way to the left I get bigger

46:05

gaps right so yeah that great way to put it um basically the stock return is sort

46:10

of more likely to make up for bad inflation this one's kind of with the

46:16

software so if we if we turn on high economic context in the historical test area does that include actual inflation

46:23

rates in the included time frame uh yes it does yep okay yeah so you'll

46:31

typically see higher higher in I know right now usually the blue you know the period in that historical graph in

46:37

encompasses the 60s and 70s at least so you'll get slightly higher inflation yeah okay so the rest are all equal

46:45

remaining questions um so kind of with the government the they said it's

46:51

important to remember that an individual's rate of inflation will be different than what the government prints homeowners will have a lower

46:58

inflation homeowners will have lower inflation than renters over the long run

47:04

yeah and I think the way to to take that into account is um you know there's several

47:10

ways you can certainly change your assumptions but uh one easy one is just to use the retirement smile or a custom

47:18

um income path meaning that um you know if the smile for example tends to reduce

47:25

real spending by like 1 to 2% a year um and another way to think about that is you

47:32

really uh just not taking full inflation adjustments you know full kind of uh you

47:39

know government printed inflation adjustments over time your personal inflation rate is

47:45

lower and then someone had put in a question and then they kind of put the question differently so putting the

47:50

question differently is not inflation one of the factors amongst others that

47:56

impact market returns and therefore Baseline infl inflation together with a variable sequence of return risk

48:03

implicitly incorporates variability in the factors impacting return including

48:09

inflation yeah I mean you basic I I I'm not maybe I don't if I'm not answering

48:14

the question directly I'm sorry I'm trying to um but I think yeah you could certainly you could you could set your

48:20

real return and and standard deviation assumptions in a way that that captured in inflation I I I really think that

48:27

that probably is possible um and so if you were able to do that um then you

48:35

would uh kind of be in a position where there's still a few of these where where there's a it's a bigger deal but if your

48:41

if your real return assumptions you know have higher um standard deviation than you might otherwise have that kind of

48:47

thing then then you might be able to capture it I think typically this gets a little bit in the weeds but I think

48:54

typically in um General planning software you're setting nominal return

48:59

assumptions and a and an inflation assumption so I'm not sure there's really a way to do that um in a typical

49:06

um software um do you see the one uh what

49:11

is the inflation adjustment for calar I don't know that accur name calsters I do

49:17

not know maybe maybe someone does know um yeah what is the inflation for calers

49:25

under California law you'll receive an automatic benefit increase equal to 2% of the initial benefit beginning September 1st after the first

49:32

anniversary of your retirement your retirement date must be before September one to receive the annual benefit

49:38

adjustment on September one of next year so I think that there's yeah you may have been responding to my example and

49:43

and to be to be clear I don't actually know whether these pensions were kers or kpers or anything like that I just remember seeing an example with an

49:50

adviser and I believe the people lived in California but I I could be misremembering um but actually it is it

49:55

is interesting to note that um you know if you have a uh something with a cola

50:02

and maybe it's two or three% yeah that'll that'll keep up with inflation if inflation is two or three% um so by

50:09

treating inflation as variable you do at least get to see that yeah this this

50:15

there there are scenarios where that's not going to be enough and it's still I still have inflation risk even though I have a um you know a cola in my in my

50:24

cash flow um so I think that's a you know it's kind of a minor point and I think people typically thought two or 3%

50:29

colas were great and then last year they started to realize oh wow yeah sometimes you might need nine

50:36

um um so everyone's been asking for the slides uh so we will be sending those

50:43

out correct Justin and and we can add those to the follow-up email when those come out with the recording of the

50:49

webinar and registration for our next webinar that'll be all in one email so we will attach the slides to that email

50:56

in addition to the registration links um I know that we are uh we don't have any more questions just as a reminder please

51:02

put your cfp ID in the survey and Justin do you have anything further that you would like to

51:08

add no that's it thank you everybody for um for joining us I hope this was uh

51:14

this was helpful I know we we try to mix in kind of these very technical um webinars with with some more practical

51:20

ones so so next month we'll be it will be a very practical one so please come back then

51:25

awesome well thank you all for joining and as always we'll see you on the next webinar and thank you for choosing

51:31

income lab bye