Total Portfolio Allocation and Withdrawal (TPAW)

[mrshikadance]

[Thread merged into here --admin LadyGeek]

Hi all,

I've been tinkering with the TPAW retirement planning tool the last few days (it's really cool) and have a few questions -

1. I've been using a ~70/30 stock bonds asset allocation in the simulations (using default returns), which I believe uses a 20 year tip return of ~1.8%. It is fair to assume this is a real yield vs nominal (I realize TIPS are inflation adjusted but just making sure)?

2.When I use the 1/CAPE simulation, I need to get very aggressive on stocks but can't get to a 70/30 simulation. Why is this?

2. Given how high the CAPE has been, I've been sitting in a 50/50 stock/cash position with the cash in a 5% money market. I'm tempted to ride this portfolio for the next few years to see how low interest rates go and to see if we get another bout of inflation. For the time being is my cash delivering a similar yield as the 20 year TIPS simulation? I believe real yields are around 2.5 - 3%.

3. I'm open to moving to 70/30 stocks/cash and I'm curious how this would affect TPAW simulations? I want to be cash heavy as this would equal ~5 years of annual consumption budget and if I retire in the next year or so, this would hopefully get me through a bear market.

For additional context, I'm in my mid-forties and am thinking of retiring the middle of next year. We have a low-seven figure portfolio with 50% in after-tax (cash) & 50% in 401k (QQQ, S&P, Gold, Intl). My wife still works and plans to till she's in her mid 50s early 60s. We live in CA and have zero debt. Monthly budget is about $8500, which TPAW simulations have shown is feasible with our portfolio size. We have a couple kids (college savings is in a separate account). We also have about $50k in an HSA.

Lastly, I'm new the forum so if I'm breaking any etiquette please let me know!

Thanks so much!

[Ben Mathew]

Ben,
Thank you for all your work on this. I have been playing around with the TPAW planner and my results are nearly identical to the Fidelity retirement planner. This is except for asset allocation. For retiring early at 49, Fidelity would have optimal AA now at 70% stocks/30 bonds. TPAW has me at 35% stock/ 65 bonds. It seems like a pretty big difference. I thought for long retirements you need higher stocks (at least 60%?). Would you help me understand the asset allocation? What am I missing?

Thank you!
DD

The asset allocation will depend on a number of factors like risk aversion, income during retirement (Social Security and pensions), and essential expenses.

But taking an example of a 49 year old retiree planning to age 95, with a $1 million portfolio and $3,000 per month of Social Security starting age 70, and risk tolerance set to the default 12, and other settings also left at their defaults, we get a median glidepath of:

- starting AA of 46/54
- rising to 83/17 by the start of Social Security
- declining to 62/38 by max age

Link to this plan: https://tpawplanner.com/link?params=jYd ... wj9vw2hvGw

You may be getting a lower stock allocation than this due to

- lower risk aversion
- higher ratio of portfolio to Social Security
- extra essential expenses (which will be funded by 100% bonds)

But generally speaking, a long retirement does not call for a higher stock allocation. Stocks are risky, and they don't become less risky over longer horizons. See Risk and Time by John Norstad about the myth of time diversification. Both the risk and the reward of holding stocks increases over time. If returns are independent, the risk and reward increase proportionally and the optimal asset allocation remains the same over all horizons. If returns are mean reverting, risk increases slower than the reward and longer horizons would merit a higher stock allocation. But even if stock returns are mean reverting, we can expect stock risk to be meaningfully higher than bonds even over long horizons. Otherwise there should be no equity risk premium over long horizons. What happens is that over long horizons, stocks become less likely to underperform bonds, but more likely to underperform more severely. For example, over the course of a day, stocks have a high probability of underperforming bonds, but a low probability of underperforming by more than 20%. Over the course of a year, stocks are less likely to underperform bonds, but more likely to underperform by more than 20%. The same effect holds over longer horizons.

The typical SWR analysis using historical returns is prone to arriving at an inappropriately stock heavy recommendation because

(1) SWR analysis grades pass/fail and ignores the severity of failure. Since stocks are less prone to underperform bonds over long horizons, but more prone to underperform severely, ignoring severity underestimates the risk of stocks.

(2) We don't have a lot of historical returns. Even 150 years of US historical returns is only three 50 year runs in a single country that has done very well during that time. Stocks are unlikely to underperform bonds in long horizons, so a small sample of returns may not contain an instance of stocks underperforming bonds over long horizons. But we can reason based on an appeal to efficient markets (and perhaps international data) that this is a rare but—importantly—not an impossible outcome that we can ignore. I wrote more about that here.

In short, we can hold a stock-heavy portfolio over long horizons if we are willing to take both the higher risk and higher return. We should not hold a stock heavy portfolio over long horizons just because we need the higher return to make the portfolio last a long time. Because with that higher return comes higher risk, which over long horizons takes the form of a small probability of severe underperformance. To make the portfolio last over a longer timeframe to support early retirement, you'd have to either save more during working years or spend less during retirement years. Stock-heavy portfolios have frequently been held up as a solution for this problem, but they are not the right solution.

[Ben Mathew]

Hi all,

I've been tinkering with the TPAW retirement planning tool the last few days (it's really cool) and have a few questions -

1. I've been using a ~70/30 stock bonds asset allocation in the simulations (using default returns), which I believe uses a 20 year tip return of ~1.8%. It is fair to assume this is a real yield vs nominal (I realize TIPS are inflation adjusted but just making sure)?

Yes, the expected return assumptions for stocks and bonds are real. The TIPS yield is the real yield obtained from "Daily Treasury Par Real Yield Curve Rates" here: https://home.treasury.gov/policy-issues ... statistics

2.When I use the 1/CAPE simulation, I need to get very aggressive on stocks but can't get to a 70/30 simulation. Why is this?

The 1/CAPE estimate for stocks is 2.7%. This is only a small premium over the expected return of bonds (TIPS yield of 1.8%). The low equity risk premium of 2.7% - 1.8% = 0.9% reduces the stock allocation per Merton's formula. So even if you increase risk tolerance to 24, you may not get to a 70/30 allocation (depending on your future cash flows).

(The default regression based estimate of the expected return of stocks is higher at 5.2% and so would imply a higher equity risk premium and a higher stock allocation.)

2. Given how high the CAPE has been, I've been sitting in a 50/50 stock/cash position with the cash in a 5% money market. I'm tempted to ride this portfolio for the next few years to see how low interest rates go and to see if we get another bout of inflation. For the time being is my cash delivering a similar yield as the 20 year TIPS simulation? I believe real yields are around 2.5 - 3%.

You can try to estimate the real yields by subtracting expected inflation rates using breakeven inflation rates available here: https://fred.stlouisfed.org/categories/33446

You may be getting high yields from money market funds, but there's no saying how long that would last. Longer term TIPS would lock in interest rates for longer.

3. I'm open to moving to 70/30 stocks/cash and I'm curious how this would affect TPAW simulations? I want to be cash heavy as this would equal ~5 years of annual consumption budget and if I retire in the next year or so, this would hopefully get me through a bear market.

I wouldn't recommend keeping a cash bucket and consuming from that to get through a bear market. Cash buckets can lead to haphazard asset allocation, not to mention a lot of confusion and arbitrary rules around when to dip in and when to replenish. Instead, keep everything in one bucket and adjust spending and asset allocation in line with the portfolio performance and expected return as calculated by TPAW. If the market drops, spending will decline. But if you are using 1/CAPE based expected returns (raw or regression based), expected returns will increase and your spending won't decline as much as the portfolio does. Stock allocation will increase if the equity risk premium has increased. The cash bucket doesn't help—it only causes confusion and mistakes.

For additional context, I'm in my mid-forties and am thinking of retiring the middle of next year. We have a low-seven figure portfolio with 50% in after-tax (cash) & 50% in 401k (QQQ, S&P, Gold, Intl). My wife still works and plans to till she's in her mid 50s early 60s. We live in CA and have zero debt. Monthly budget is about $8500, which TPAW simulations have shown is feasible with our portfolio size. We have a couple kids (college savings is in a separate account). We also have about $50k in an HSA.

I'd go with longer term TIPS instead of money market to better match your spending duration. Also reconsider QQQ—it's a risky bet on tech stocks.

[LadyGeek]

I merged mrshikadance's thread into the ongoing discussion.

(Thanks to the member who reported the post and explained what's wrong.)

[mrshikadance]

I wouldn't recommend keeping a cash bucket and consuming from that to get through a bear market. Cash buckets can lead to haphazard asset allocation, not to mention a lot of confusion and arbitrary rules around when to dip in and when to replenish. Instead, keep everything in one bucket and adjust spending and asset allocation in line with the portfolio performance and expected return as calculated by TPAW. If the market drops, spending will decline. But if you are using 1/CAPE based expected returns (raw or regression based), expected returns will increase and your spending won't decline as much as the portfolio does. Stock allocation will increase if the equity risk premium has increased. The cash bucket doesn't help—it only causes confusion and mistakes.

So amazingly helpful. Appreciate you, Ben! A few more questions:

1) In term of buying TIPS, I'm assuming you mean buying a TIPS ladder vs an ETF

2) If so, would one go about the practicalities of paying themselves a salary/having cash for monthly expenses by combining the interest payments with stock asset sales? I guess, I just always assumed folks need to keep a few years of cash on hand to manage expenses and then those funds would get topped up every so often.

3) It sounds like TPAW adjusts asset allocation over time based on market conditions/portfolio performance and glide path. Is there a way to see the projected glide path simulations? Additionally, would a user at some point hit "go" at which point they would log in every month to see their budget and needed allocation changes?

4) In regard to your comment about me using 1/CAPE and the spending budget not declining as much as the portfolio, is the same true if I'm using Regression Prediction (~5% expected returns)?

[dachshunddad]

Ben,
Thank you for all your work on this. I have been playing around with the TPAW planner and my results are nearly identical to the Fidelity retirement planner. This is except for asset allocation. For retiring early at 49, Fidelity would have optimal AA now at 70% stocks/30 bonds. TPAW has me at 35% stock/ 65 bonds. It seems like a pretty big difference. I thought for long retirements you need higher stocks (at least 60%?). Would you help me understand the asset allocation? What am I missing?

Thank you!
DD

The asset allocation will depend on a number of factors like risk aversion, income during retirement (Social Security and pensions), and essential expenses.

But taking an example of a 49 year old retiree planning to age 95, with a $1 million portfolio and $3,000 per month of Social Security starting age 70, and risk tolerance set to the default 12, and other settings also left at their defaults, we get a median glidepath of:

- starting AA of 46/54
- rising to 83/17 by the start of Social Security
- declining to 62/38 by max age

Link to this plan: https://tpawplanner.com/link?params=jYd ... wj9vw2hvGw

You may be getting a lower stock allocation than this due to

- lower risk aversion
- higher ratio of portfolio to Social Security
- extra essential expenses (which will be funded by 100% bonds)

But generally speaking, a long retirement does not call for a higher stock allocation. Stocks are risky, and they don't become less risky over longer horizons. See Risk and Time by John Norstad about the myth of time diversification. Both the risk and the reward of holding stocks increases over time. If returns are independent, the risk and reward increase proportionally and the optimal asset allocation remains the same over all horizons. If returns are mean reverting, risk increases slower than the reward and longer horizons would merit a higher stock allocation. But even if stock returns are mean reverting, we can expect stock risk to be meaningfully higher than bonds even over long horizons. Otherwise there should be no equity risk premium over long horizons. What happens is that over long horizons, stocks become less likely to underperform bonds, but more likely to underperform more severely. For example, over the course of a day, stocks have a high probability of underperforming bonds, but a low probability of underperforming by more than 20%. Over the course of a year, stocks are less likely to underperform bonds, but more likely to underperform by more than 20%. The same effect holds over longer horizons.

The typical SWR analysis using historical returns is prone to arriving at an inappropriately stock heavy recommendation because

(1) SWR analysis grades pass/fail and ignores the severity of failure. Since stocks are less prone to underperform bonds over long horizons, but more prone to underperform severely, ignoring severity underestimates the risk of stocks.

(2) We don't have a lot of historical returns. Even 150 years of US historical returns is only three 50 year runs in a single country that has done very well during that time. Stocks are unlikely to underperform bonds in long horizons, so a small sample of returns may not contain an instance of stocks underperforming bonds over long horizons. But we can reason based on an appeal to efficient markets (and perhaps international data) that this is a rare but—importantly—not an impossible outcome that we can ignore. I wrote more about that here.

In short, we can hold a stock-heavy portfolio over long horizons if we are willing to take both the higher risk and higher return. We should not hold a stock heavy portfolio over long horizons just because we need the higher return to make the portfolio last a long time. Because with that higher return comes higher risk, which over long horizons takes the form of a small probability of severe underperformance. To make the portfolio last over a longer timeframe to support early retirement, you'd have to either save more during working years or spend less during retirement years. Stock-heavy portfolios have frequently been held up as a solution for this problem, but they are not the right solution.

Thanks for taking the time to explain it in detail. That makes a lot of sense. I appreciate you unpacking why a longer retirement doesn’t necessarily mean higher stock allocation.

[Ben Mathew]

1) In term of buying TIPS, I'm assuming you mean buying a TIPS ladder vs an ETF

Not necessarily. The duration of the bonds is what matters. ETFs holding TIPS are fine if you can get the right duration. Ladders work best when you are creating a floor and so don't have to rebalance the bonds. When the TIPS are part of the the risk portfolio and need to be rebalanced, it's more convenient to hold ETFs that you can easily buy and sell to rebalance to the target asset allocation.

2) If so, would one go about the practicalities of paying themselves a salary/having cash for monthly expenses by combining the interest payments with stock asset sales? I guess, I just always assumed folks need to keep a few years of cash on hand to manage expenses and then those funds would get topped up every so often.

Since you'll be rebalancing to a target asset allocation, you'll be consuming from the overweight asset—whether that's stocks or bonds. If the withdrawal amount is greater than interest payments and dividends, you'll have to sell stocks and/or bonds depending on which assets are overweight and how much you're withdrawing. Exactly how that plays out is determined by the spending and asset allocation targets, and it's usually simpler to think about it in those terms rather than as consuming out of interest, dividends, stock sales, or bond sales. You just need to spend X and rebalance to Y.

3) It sounds like TPAW adjusts asset allocation over time based on market conditions/portfolio performance and glide path. Is there a way to see the projected glide path simulations?

Yes. Click on the dropdown next to the title of the graph in the results panel and select "Asset Allocation." This will bring up the graph showing the simulated asset allocation over time.

Additionally, would a user at some point hit "go" at which point they would log in every month to see their budget and needed allocation changes?

Yes, once you've arrived at a plan that you most prefer, you can go into implementation mode. Periodically—maybe once a month or so—enter your current portfolio balance and see the updated asset allocation and contribution/withdrawal amounts listed in "Tasks" and implement that (while taking care not to rebalance too frequently.) Less frequently—maybe once or twice a year or after major life changes—review and update the inputs in the plan.

4) In regard to your comment about me using 1/CAPE and the spending budget not declining as much as the portfolio, is the same true if I'm using Regression Prediction (~5% expected returns)?

Yes, the "Regression Prediction" estimate of the expected stock return will move in the same direction as 1/CAPE. So, like 1/CAPE, it will be countercyclical and cause spending to decline by less than the decline in the portfolio.

[Raspberry-503]

i don't know if this should be in its own thread, but I'm one of those people that learns best by doing. I'm making a spreadsheet that tries to calculate/emulate some of the TPAW calculations because that's how I will really understand them.
The wiki has TPAW spreadsheets that I forgot about so I ended up reinventing the wheel a little bit but now comparing spreadsheet to the TPAW tool.
One thing I had never noticed before was the "Merton's input matched to TPAW inputs" spreadsheet. The good news is that the derived allocation matches my own calculations using Merton's Share formula

Playing around with those, I think I may now know why my spreadsheet yields a higher recommend withdrawal (looking at year/month 1 only).
Setting up the simplest possible case"
- Retired today at age 56
- Plan to 100 (44 years)
- Portfolio today is $1.5M
- No SS or other supplemental income ever
- Real stock return 5.2% and Real bond (safe and risky) return 1.7%
- RRA of 2.26 (I used TPAW for that and picked the highest Moderate setting)
- No legacy

All sliders in TPAW are set to default except for the legacy-risk one which is set to max since I don't want to leave a legacy in this example

The Merton equation shows AA should be 48/52, at that AA the portfolio return rate is 3.4%, that's also what TPAW says of course

With such simple inputs PV of the portfolio is $1.5M, 44 years of 528 months (my spreadsheet uses months)
My spreadsheet returns PMT ( (1-3.4)^(1/12)-1, 528 months, $1.5M, 0,1 ) = $5417/month or about $65,000/year

TPAW says $3,766 or about $45,200 per year

The TPAW retiree spreadsheet from the wiki (Monte Carlo version but irrelevant here) says $45,453/year when loaded with the Merton to TPAW parameters

So TPAW and its spreadsheet agree, but are much lower than my spreadsheet

But the Merton to TPAW spreadsheet told me to enter g=1.97% (growth rate of withdrawals). My PMT formula as far as I understand it has a g of 0, in fact I don't know how to account for g in PMT, I only know how to do it with a table where g is applied to the previous line

If I set g=0 in the TPAW spreadsheet, I get a yearly withdrawal of $63,800, much closer to my calculation. I think the differences may be fencepost issues and year vs month calculations.

Armed with that knowledge I now notice that even though I set my spending tilt to 0 in TPAW, it doesn't mean g=0, if I expand "spending tile breakdown" it shows that TPAW automatically added 1.96% that is "calculated for you based on your risk tolerance and preference for the future"

How about that!, it's (basically) the same as the g recommended by the Merton to TPAW calculator.

Conversely if I set "Preference for the future" in TPAW at -4.%, the total tilt is now 0 and month 1 withdrawal is within a few dollars of my calculation.

The spreadsheet seems to be computing g from two successive consumption calculations using Merton's consumption formula, so Merton's consumption formula is an amortization with a g which itself depends depends in gamma/RRA. Lightbulb! Up until now I thought that I could calculate withdrawals with a simple PMT calculation.

I need to ponder this, I want to favor early retirement so in my mind was 0 if not negative, but now the Monte Carlo shows the baseline to steadily drop down when I would have expected it to be flat as long as the portfolio returns the same consistent return. So the Monte Carlo simulation indicates that the median historical return is less that 3.4% for the portfolio?

[Raspberry-503]

I'm still reading a post on page 3 of this thread where Ben explains Merton is ABW with a non-zero g
The Merton to TPAW sheet has a time discount value that I left at 0 that plays into this too

[Ben Mathew]

i don't know if this should be in its own thread, but I'm one of those people that learns best by doing. I'm making a spreadsheet that tries to calculate/emulate some of the TPAW calculations because that's how I will really understand them.
The wiki has TPAW spreadsheets that I forgot about so I ended up reinventing the wheel a little bit but now comparing spreadsheet to the TPAW tool.
One thing I had never noticed before was the "Merton's input matched to TPAW inputs" spreadsheet. The good news is that the derived allocation matches my own calculations using Merton's Share formula

Playing around with those, I think I may now know why my spreadsheet yields a higher recommend withdrawal (looking at year/month 1 only).
Setting up the simplest possible case"
- Retired today at age 56
- Plan to 100 (44 years)
- Portfolio today is $1.5M
- No SS or other supplemental income ever
- Real stock return 5.2% and Real bond (safe and risky) return 1.7%
- RRA of 2.26 (I used TPAW for that and picked the highest Moderate setting)
- No legacy

All sliders in TPAW are set to default except for the legacy-risk one which is set to max since I don't want to leave a legacy in this example

The Merton equation shows AA should be 48/52, at that AA the portfolio return rate is 3.4%, that's also what TPAW says of course

With such simple inputs PV of the portfolio is $1.5M, 44 years of 528 months (my spreadsheet uses months)
My spreadsheet returns PMT ( (1-3.4)^(1/12)-1, 528 months, $1.5M, 0,1 ) = $5417/month or about $65,000/year

TPAW says $3,766 or about $45,200 per year

The TPAW retiree spreadsheet from the wiki (Monte Carlo version but irrelevant here) says $45,453/year when loaded with the Merton to TPAW parameters

So TPAW and its spreadsheet agree, but are much lower than my spreadsheet

But the Merton to TPAW spreadsheet told me to enter g=1.97% (growth rate of withdrawals). My PMT formula as far as I understand it has a g of 0, in fact I don't know how to account for g in PMT, I only know how to do it with a table where g is applied to the previous line

If I set g=0 in the TPAW spreadsheet, I get a yearly withdrawal of $63,800, much closer to my calculation. I think the differences may be fencepost issues and year vs month calculations.

Armed with that knowledge I now notice that even though I set my spending tilt to 0 in TPAW, it doesn't mean g=0, if I expand "spending tile breakdown" it shows that TPAW automatically added 1.96% that is "calculated for you based on your risk tolerance and preference for the future"

How about that!, it's (basically) the same as the g recommended by the Merton to TPAW calculator.

Conversely if I set "Preference for the future" in TPAW at -4.%, the total tilt is now 0 and month 1 withdrawal is within a few dollars of my calculation.

The spreadsheet seems to be computing g from two successive consumption calculations using Merton's consumption formula, so Merton's consumption formula is an amortization with a g which itself depends depends in gamma/RRA. Lightbulb! Up until now I thought that I could calculate withdrawals with a simple PMT calculation.

I'm still reading a post on page 3 of this thread where Ben explains Merton is ABW with a non-zero g
The Merton to TPAW sheet has a time discount value that I left at 0 that plays into this too

Yes, the difference between your calculations and TPAW calculations are due to the base spending tilt implied by Merton's formula. The spending tilt slider is an "extra" spending tilt that can be used to add or subtract from the base spending tilt.

The base spending tilt comes from the "preference for the future" (which corresponds to the "time preference rate.") The higher your preference for the future, the higher the base spending tilt will be. But even with preference for the future set to 0% (the default setting), you'll still get a positive base spending tilt for the following reasons:

From https://tpawplanner.com/learn/mertons-portfolio-problem:

But why would anyone want to schedule rising or declining spending rather than flat spending like Sally did? There are a few reasons:

1. Precautionary savings: This is saving for a rainy day. The distant future is more uncertain than the near future. More time creates more opportunity for life to deviate from expectations. Unforeseen spending needs accumulate. Portfolio gains and losses multiply. So your circumstances twenty or thirty years from now is likely to be quite different—either much better or much worse—than your circumstances two or three years from now. So caution dictates that you spend less early on than you otherwise might have, just in case things turn out badly later. This instinct to save in early retirement to protect against the uncertainties of late retirement is called precautionary savings.

2. The future is cheaper: Since the portfolio is growing, consuming later is cheaper. If you give up $1 in early retirement, you’ll get $1.50 in late retirement. This may persuade some people to delay their spending to increase the total amount that they can spend.

More on the extra spending tilt vs the base spending tilt here.

I need to ponder this, I want to favor early retirement so in my mind was 0 if not negative, but now the Monte Carlo shows the baseline to steadily drop down when I would have expected it to be flat as long as the portfolio returns the same consistent return. So the Monte Carlo simulation indicates that the median historical return is less that 3.4% for the portfolio?

This is because the median outcome will be below the expected outcome due to volatility drag. If you make the total spending tilt 0% (by reducing the preference for the future and/or the extra spending tilt), the expected spending will be constant. If the market does exactly as expected (3.4% in the above example) each year, spending will stay flat. But the median outcome will be below that since it comes from a mix of returns. So the median spending will be lower than flat—i.e. it will be declining.

Example of expected vs median outcomes: Suppose returns are drawn randomly from {5%, 10%, 15%} with equal probability. Then:

• Expected return = simple average of {5%, 10%, 15%} = 10.00%

• Expected balance after $1 is invested for 3 periods = (1+.10)((1+.10)(1+.10) = $1.331. [Geometric mean return: 10.00%]

• Median balance after $1 is invested for 3 periods = (1+.05)((1+.10)(1+.15) = $1.328. [Geometric mean return: 9.92%]

The median balance comes from a mix of returns and so will end up lower than the expected balance.

If you want to make the median spending stay constant, you'd have to increase the spending tilt to cover the gap between the expected and median return. In this example, that would mean amortizing the portfolio using an amortization rate of expected return of 10% - spending tilt of 0.08% = 9.92% (the geometric mean of the median outcome.)

[Raspberry-503]

Thanks Ben as always...

[Raspberry-503]

I found this blog from Boglehead forum user Siamond :https://www.bogleheads.org/blog/2019/02 ... -of-money/

My spreadsheet is actually a lot closer to what he is doing there than to the TPAW spreadsheet, even the Savings Portfolio version. I had not appreciated some of the subtleties built into the TPAW spreadsheets (and the online tool), especially the Total Portfolio version.

The commonality is in treating present value of future cash flows as part of today's portfolio with a relevant discount rate. The main difference as explained above is that I have been using a straight PMT formula to calculate the withdrawal, so g=0, and of course no rebalancing of the risk portfolio as is done with TPAW total portfolio.

The results are pretty close in the case of my real-life retirement plan, my "I'm retired today" scenario has about 2% deviation between my spreadsheet and TPAW's recommendation for this month's withdrawal in Savings Portfolio mode and 7% from Total Portfolio mode. It's pretty easy to construct simple examples where it deviates more, like in the example above my spreadsheet is 40% higher than TPAW.

[Ben Mathew]

I found this blog from Boglehead forum user Siamond :https://www.bogleheads.org/blog/2019/02 ... -of-money/

My spreadsheet is actually a lot closer to what he is doing there than to the TPAW spreadsheet, even the Savings Portfolio version. I had not appreciated some of the subtleties built into the TPAW spreadsheets (and the online tool), especially the Total Portfolio version.

The commonality is in treating present value of future cash flows as part of today's portfolio with a relevant discount rate. The main difference as explained above is that I have been using a straight PMT formula to calculate the withdrawal, so g=0, and of course no rebalancing of the risk portfolio as is done with TPAW total portfolio.

The results are pretty close in the case of my real-life retirement plan, my "I'm retired today" scenario has about 2% deviation between my spreadsheet and TPAW's recommendation for this month's withdrawal in Savings Portfolio mode and 7% from Total Portfolio mode. It's pretty easy to construct simple examples where it deviates more, like in the example above my spreadsheet is 40% higher than TPAW.

Siamond's approach uses the expected return of the portfolio as the discount rate to calculate the present value (PV) of future cash flows. TPAW uses the bond rate as the discount rate for income during retirement and essential extra expenses, and the expected return of the portfolio as the discount rate for discretionary extra expenses. Siamond's approach adds the PV of net future cash flows to the portfolio for the withdrawal calculation, not for asset allocation (asset allocation is predetermined and fixed.) TPAW adds the PV of net future cash flows for both asset allocation and withdrawal calculations.

There was a long discussion about the appropriate use of discount rates in the ABW thread when we were writing the ABW wiki a few years ago. To read that discussion, you can start with this post on page 7 and read till this post on page 11 where we agree to disagree and present both approaches in the wiki.

In the ABW wiki, Siamond's approach is described in the savings portfolio focused approach section and the TPAW approach is described in the total portfolio focused approach section.

In the online TPAW planner, these methods are implemented as the SPAW and TPAW strategies respectively.

[Raspberry-503]

Thanks for explaining the differences Ben, that matches my understanding now after comparing both. That's why I refer to TPAW in "Total Portfolio Mode " and "Savings Portfolio mode" in my previous post.
For my own spreadsheet I use the TIPS rate for guarantied cash flows like SS, TIPS ladder, ... and the portfolio rate for everything else (say travel expenses) since they are discretionary, so I guess it's closer to TPAW (SPAW) in that regard. And yes, working through my spreadsheet helped me understand Total vs Savings rebalancing better and how/why there is a tilt towards future consumption (g>0). While you don't need to understand all of this to use TPAW, it makes me feel much more confident that it's a great and flexible tool.

[mrshikadance]

Thanks so much Ben! A few final questions for you -

Given that I'm in 40s, I've previously never been a big bond buyer but it's super helpful to understand how bonds can help with wealth preservation/safety (and some yield relative to cash) as I plan for an early retirement (in a couple of years).

TPAW suggests that I do a 50/50 bond/stock split. I plan to go 100% bonds in my after-tax account and use my 401K for 100% stocks, which nets out to ~50/50.

I did a bit of math on TIPS vs nominal bonds and it looks like TIPS are slightly ahead on 3 year/5 year basis (I used the Phili Fed ten year inflation projections). Given that these assets take interest rate risk of the table, I'm far more inclined to go with them vs ETFs like BND/SCHP. My thinking is that these assets are subject to market vol/changing economic/market dynamics and the risk/reward given their long run returns aren't worth it. Anything I'm missing?

1) In term of buying TIPS, I'm assuming you mean buying a TIPS ladder vs an ETF

Not necessarily. The duration of the bonds is what matters. ETFs holding TIPS are fine if you can get the right duration. Ladders work best when you are creating a floor and so don't have to rebalance the bonds. When the TIPS are part of the the risk portfolio and need to be rebalanced, it's more convenient to hold ETFs that you can easily buy and sell to rebalance to the target asset allocation.

2) If so, would one go about the practicalities of paying themselves a salary/having cash for monthly expenses by combining the interest payments with stock asset sales? I guess, I just always assumed folks need to keep a few years of cash on hand to manage expenses and then those funds would get topped up every so often.

Since you'll be rebalancing to a target asset allocation, you'll be consuming from the overweight asset—whether that's stocks or bonds. If the withdrawal amount is greater than interest payments and dividends, you'll have to sell stocks and/or bonds depending on which assets are overweight and how much you're withdrawing. Exactly how that plays out is determined by the spending and asset allocation targets, and it's usually simpler to think about it in those terms rather than as consuming out of interest, dividends, stock sales, or bond sales. You just need to spend X and rebalance to Y.

3) It sounds like TPAW adjusts asset allocation over time based on market conditions/portfolio performance and glide path. Is there a way to see the projected glide path simulations?

Yes. Click on the dropdown next to the title of the graph in the results panel and select "Asset Allocation." This will bring up the graph showing the simulated asset allocation over time.

Additionally, would a user at some point hit "go" at which point they would log in every month to see their budget and needed allocation changes?

Yes, once you've arrived at a plan that you most prefer, you can go into implementation mode. Periodically—maybe once a month or so—enter your current portfolio balance and see the updated asset allocation and contribution/withdrawal amounts listed in "Tasks" and implement that (while taking care not to rebalance too frequently.) Less frequently—maybe once or twice a year or after major life changes—review and update the inputs in the plan.

4) In regard to your comment about me using 1/CAPE and the spending budget not declining as much as the portfolio, is the same true if I'm using Regression Prediction (~5% expected returns)?

Yes, the "Regression Prediction" estimate of the expected stock return will move in the same direction as 1/CAPE. So, like 1/CAPE, it will be countercyclical and cause spending to decline by less than the decline in the portfolio.

[Ben Mathew]

I did a bit of math on TIPS vs nominal bonds and it looks like TIPS are slightly ahead on 3 year/5 year basis (I used the Phili Fed ten year inflation projections). Given that these assets take interest rate risk of the table, I'm far more inclined to go with them vs ETFs like BND/SCHP. My thinking is that these assets are subject to market vol/changing economic/market dynamics and the risk/reward given their long run returns aren't worth it. Anything I'm missing?

I wouldn't put too much weight on TIPS having a slightly higher expected real yield than nominal bonds based on a particular inflation estimate. Inflation estimates are too imprecise to tell for sure. The important thing is that the difference in expected real yields are close enough to zero that people can argue about whether it's positive or negative. In fact, the breakeven inflation rate estimate of expected inflation is based on the assumption that the difference is zero.

If there isn't a significant premium for TIPS over nominal bonds (and there doesn't seem to be), it would be better to go with TIPS to remove inflation risk. The reason TIPS (and real bonds in general) are not more popular than nominal bonds is likely money illusion.

The other source of bond risk (besides default, which is negligible for US government bonds) is the risk of changing real interest rates. This interest rate risk can be reduced (if the bonds are part of the risk portfolio and so will be rebalanced) or eliminated (if the bonds are part of the spending floor and so will not be rebalanced) by duration matching the TIPS to spending horizons. Whether the TIPS are held directly or through an ETF does not matter—what matters is their duration.

[GAAP]

I did a bit of math on TIPS vs nominal bonds and it looks like TIPS are slightly ahead on 3 year/5 year basis (I used the Phili Fed ten year inflation projections). Given that these assets take interest rate risk of the table, I'm far more inclined to go with them vs ETFs like BND/SCHP. My thinking is that these assets are subject to market vol/changing economic/market dynamics and the risk/reward given their long run returns aren't worth it. Anything I'm missing?

I wouldn't put too much weight on TIPS having a slightly higher expected real yield than nominal bonds based on a particular inflation estimate. Inflation estimates are too imprecise to tell for sure. The important thing is that the difference in expected real yields are close enough to zero that people can argue about whether it's positive or negative. In fact, the breakeven inflation rate estimate of expected inflation is based on the assumption that the difference is zero.

If there isn't a significant premium for TIPS over nominal bonds (and there doesn't seem to be), it would be better to go with TIPS to remove inflation risk. The reason TIPS (and real bonds in general) are not more popular than nominal bonds is likely money illusion.

The other source of bond risk (besides default, which is negligible for US government bonds) is the risk of changing real interest rates. This interest rate risk can be reduced (if the bonds are part of the risk portfolio and so will be rebalanced) or eliminated (if the bonds are part of the spending floor and so will not be rebalanced) by duration matching the TIPS to spending horizons. Whether the TIPS are held directly or through an ETF does not matter—what matters is their duration.

And someone in their 40s likely has a duration need for retirement well in excess of that available from mid-term bonds. I would start with a planning need of:

• Duration = (retirement age - current age) + 0.5 * (life expectancy - retirement age)

IRS RMD Table 1 will give a starting point estimate for life expectancy. At age 45, with a planned retirement at 65, that would yield 20 + 0.5*41 = about 40 years. NOTE: this is not necessarily a popular position here...

[clobber88]

Conservative: Risk tolerance = 7 (RRA = 6.48)
Moderate: Risk tolerance = 12 (RRA = 3.05)
Aggressive: Risk tolerance = 17 (RRA = 1.44)

Hey Ben,

A few quick questions. I've been using the tool for a while and have posted a few times.

1) Is there a tool to try and gauge Risk Tolerance?
2) Is there a tool to try and gauge RRA?
3) If TPAW is using RRA under the hood, why use Risk Tolerance for the input?

thanks

[Raspberry-503]

I find that risk tolerance is generally understood (misunderstood?) when expressed as a range from most conservative to most aggressive. One could argue about whether the tools and questionnaires out there (from brokers and advisors) work, but as poor as a measure as it is, it probably is what people will relate to. So I know in my case I appreciated being able to dial how aggressive or not I want to be, and have the tool translate it to an RRA.

The Monte Carlo analysis that TPAW does is another way to guide your decision. Look at the min max a few years out, way out into retirement. Can you put up with the bottom number? Does the top number look like you'll never find a way to spend this much money? Play with the slider until you find Monte Carlo results that you can live with. That gives you an idea of both your risk tolerance and your risk capacity (aka, if you won the game stop playing)

[mrshikadance]

Thanks again, Ben. I'm following everything you're saying except for the ETF and directly held bonds being functionally the same. If I buy an ETF wit a 6 year avg duration and build a ladder with the essentially the same duration, isn't it possible that the ETF yield (price) in 6 years is lower/higher than the bond ladder with the inflation protected interest payments?

The other source of bond risk (besides default, which is negligible for US government bonds) is the risk of changing real interest rates. This interest rate risk can be reduced (if the bonds are part of the risk portfolio and so will be rebalanced) or eliminated (if the bonds are part of the spending floor and so will not be rebalanced) by duration matching the TIPS to spending horizons. Whether the TIPS are held directly or through an ETF does not matter—what matters is their duration.

[ConstantChrysalis]

Thanks again, Ben. I'm following everything you're saying except for the ETF and directly held bonds being functionally the same. If I buy an ETF wit a 6 year avg duration and build a ladder with the essentially the same duration, isn't it possible that the ETF yield (price) in 6 years is lower/higher than the bond ladder with the inflation protected interest payments?

The other source of bond risk (besides default, which is negligible for US government bonds) is the risk of changing real interest rates. This interest rate risk can be reduced (if the bonds are part of the risk portfolio and so will be rebalanced) or eliminated (if the bonds are part of the spending floor and so will not be rebalanced) by duration matching the TIPS to spending horizons. Whether the TIPS are held directly or through an ETF does not matter—what matters is their duration.

I believe that Ben is assuming using two bond ETFs in a ratio where the weighted average of the ETF durations is 1/2 of the target timeframe (in this case, ladder length), rebalanced about quarterly as the timeframe shortens.

[ConstantChrysalis]

Conservative: Risk tolerance = 7 (RRA = 6.48)
Moderate: Risk tolerance = 12 (RRA = 3.05)
Aggressive: Risk tolerance = 17 (RRA = 1.44)

1) Is there a tool to try and gauge Risk Tolerance?

The book "Missing Billionaires" has an extensive discussion on this topic and has a self-assessment quiz to find your RRA. Also, Harvard's John Campbell uses this thought experiment:

"Imagine a demon enters your life and insists you gamble 20% of your wealth on the flip of a fair coin. If it lands on heads your wealth increases by 20% and if it lands on tails your wealth is reduced by 20%. What is the largest percentage of your wealth you would be willing to pay the demon to make him go away and not make you take the gamble? Your gamma is roughly equal to that percentage divided by 2. So, if you would pay 4% of your wealth to avoid the gamble, you have a risk-aversion of 2."

The "Missing Billionaires" authors surveyed their colleagues using their instrument and most were a 3, some were more risk tolerant at 2 and very few at 1. On the risk-adverse side, about the same number who were a 3 were split between 4 and 5. The pool was, however, filled with financially savvy and wealthy people.

Ben counsels against using the results of these tools blindly in the TPAW planner and instead encourages experimenting with the Risk Tolerance slider until you find a spending profile that is most desirable for you.

[GAAP]

The best tool I've found to measure risk tolerance is experience -- which unfortunately tends to apply when you overestimate.

I've also found that risk in general while accumulating is different from risk when retired. Tolerance level may change when that change actually occurs -- and may be different from what was planned or expected.

Portfolio withdrawals are large drags that increase drawdowns and slow drawdown recovery. That can have a fairly dramatic effect on the perception of portfolio performance and thus the perception of actualized risk.

[Raspberry-503]

I would have no idea how to deal with that demon. I guess I'm not smart enough for the thought experiment.
I read The Missing Billionaires so I knew the median was 3, which is also what you get in TPAW when you set it in the middle, with the highest "moderate" setting in TPAW having RRA=2.26 (risk tolerance 14/25) and the lowest has RRA=4.12 (risk tolerance 10/25)
For each RRA you get a different recommended allocation, but to most people approaching TPAW the "low" equity lalocation for the total portfolio is something a lot harder to wrap your head around than deciding your RRA.

[Ben Mathew]

Conservative: Risk tolerance = 7 (RRA = 6.48)
Moderate: Risk tolerance = 12 (RRA = 3.05)
Aggressive: Risk tolerance = 17 (RRA = 1.44)

Hey Ben,

A few quick questions. I've been using the tool for a while and have posted a few times.

1) Is there a tool to try and gauge Risk Tolerance?
2) Is there a tool to try and gauge RRA?
3) If TPAW is using RRA under the hood, why use Risk Tolerance for the input?

thanks

Risk tolerance and relative risk aversion (RRA) are the same thing, just expressed in different ways. A risk tolerance of 12 corresponds to an RRA of 3.05, 13 corresponds to 2.62, etc.

We converted RRA to risk tolerance and used that on the slider because it leads to a more intuitive slider. The left side of a risk spectrum is usually conservative and the right side aggressive. If we put RRA directly on the slider, that would become inverted: left would be low RRA and aggressive, and right would be high RRA and conservative, which would be confusing. When converting RRA to risk tolerance, we also use a log scale because the difference between an RRA of 1 and 2 is greater than the difference between an RRA of 5 and 6. The log scale spreads it out better for the slider.

The best way to gauge your risk tolerance/RRA is by simply adjusting the risk tolerance slider till you arrive at the spending graph (and legacy outcomes) that you most prefer. There are tools like surveys and thought experiments to help you evaluate your RRA, but those are less reliable because you are not making the evaluation in the context of real world financial circumstances. They are based on artificial scenarios that may not translate well enough to your actual circumstance and don't provide the full context of the choices that you are facing. This is especially a problem if you have created a spending floor (by adding essential extra expenses). Your risk aversion should be evaluated in the context of that floor. If you have a floor that covers your essential expenses, you can take more risk on the risk portfolio. That would mean entering a lower RRA (higher risk tolerance) on the slider. A survey can miss this if you're not careful to account for the floor when answering questions. But the planner will catch it. Evaluating your risk tolerance inside the planner by adjusting the risk tolerance slider will avoid these types of errors.

RRA simply measures how willing you are to take on risk. That risk is displayed in the spending graph. You don't need anything more. If you get an RRA from a survey, that is simply a guess about how you would adjust the risk slider when faced with these choices. You don't need that guess—you can just adjust the slider and find it directly!

If you still want to get an RRA estimate from outside the planner, the Missing Billionaires survey (pages 175-180) and Professor Campbell's demon thought experiment that ConstantChrysalis referred to above are good sources. But make sure that you don't blindly enter the results in the planner. Instead, look at the spending graph in the planner and adjust the risk aversion slider till you find the spending distribution that you most prefer. If the RRA you get from the planner is different from the survey and thought experiment results, go with the RRA you found through the planner. It's effectively a richer survey administered in the perfect context—the financial choices that you actually face in real life.

Note also that this approach of adjusting the inputs based on the outcomes shown in the spending graph is how the planner is meant to be used for all inputs, not just for risk tolerance. For example, the choices of how much to save, when to retire, etc, should be made based on how it’s affecting the spending graph. The spending graph is what captures the material impact of all of those decisions on retirement. This is why it's the default view of the planner and why we have taken care to make the graph visible when you enter any of the inputs. It always boils down to what's happening in the spending graph.

[Ben Mathew]

Thanks again, Ben. I'm following everything you're saying except for the ETF and directly held bonds being functionally the same. If I buy an ETF wit a 6 year avg duration and build a ladder with the essentially the same duration, isn't it possible that the ETF yield (price) in 6 years is lower/higher than the bond ladder with the inflation protected interest payments?

The other source of bond risk (besides default, which is negligible for US government bonds) is the risk of changing real interest rates. This interest rate risk can be reduced (if the bonds are part of the risk portfolio and so will be rebalanced) or eliminated (if the bonds are part of the spending floor and so will not be rebalanced) by duration matching the TIPS to spending horizons. Whether the TIPS are held directly or through an ETF does not matter—what matters is their duration.

As ConstantChrysalis noted above, matching the bond ladder duration with bond ETFs would involve rebalancing from longer duration to shorter duration ETFs as the ladder horizon shrinks.

For example, a 12 year ladder will have about a 6 year duration to begin with. (It would actually be slightly less than 6 years because it should be the present value weighted average, not a simple average. But let's ignore that for this example.)

To replicate this ladder using ETFs, you would hold a mix of ETFs with an average duration of 6 years. If the interest rate changes, the change in the value of the ETF portfolio will be the same as the change in the value of the TIPS ladder. So regardless of how the interest rate moves, you can fund the same spending with the ETFs as you can with the TIPS ladder.

A year later, the bond ladder goes out only 11 years and so will have a duration of about 5.5 years. To replicate that, now you'll need to hold a mix of ETFs with an average duration of 5.5 years. So you'd rebalance your bond portfolio to get the average duration down from 6 years to 5.5 years. That means rebalancing from a longer duration ETF to a shorter duration ETF. After you do that, again, if the interest rate changes, the change in the value of the ETF portfolio will match the change in the value of the TIPS ladder.

So as long as you keep rebalancing like this, each year the bond returns will match and you can get the same spending outcome using either strategy. So by matching a 12 year bond ladder with ETFs, you can consume what the ladder would have paid out for 12 years and be left with the same $0 balance at the end.

Duration matching using ETFs won't work perfectly because the yield curve is not flat and different parts of the yield curve can shift differently. For example, if long term rates go up only a little while short term rates go up a lot, the returns of the ETF portfolio and the bond ladder will diverge. Exact duration matching would require creating the full ladder. But holding a mix of ETFs with the same average duration as the ladder may be more convenient when you need to buy and sell stocks to rebalance along the way (as you would in a risk portfolio) and may be a good enough alternative.

[corn18]

Just wanted to make my occasional post expressing my gratitude for the TPAW model. SO super simple and easy to model and provides clarity that other models do not. Thank you Ben and everyone that makes TPAW great!

[Raspberry-503]

But holding a mix of ETFs with the same average duration as the ladder may be more convenient when you need to buy and sell stocks to rebalance along the way (as you would in a risk portfolio) and may be a good enough alternative.

Also holding ETFs allows you to automatically reinvest the interest kicked off by the bonds instead of having to manually reinvest.

[GAAP]

But holding a mix of ETFs with the same average duration as the ladder may be more convenient when you need to buy and sell stocks to rebalance along the way (as you would in a risk portfolio) and may be a good enough alternative.

Also holding ETFs allows you to automatically reinvest the interest kicked off by the bonds instead of having to manually reinvest.

It also far easier to extend duration if the need arises when using ETFs.

I have an open-ended longevity estimate that I duration match to. As I age, the duration decreases, but it does so at a slower rate. Right now, I may need a 20 year duration, but f I'm alive in 20 years I would still need something greater than zero. Achieving that with ETFs simply requires a slight change in the relative proportion whereas a ladder would require the purchase of additional rungs.

IMO, ladders work best for fixed timeframes -- deferred Social Security is probably the biggest/best example. If change is more likely, ETF flexibility is a better fit.

Others here disagree -- sometimes fairly vehemently.

[ScubaHogg]

Just wanted to make my occasional post expressing my gratitude for the TPAW model. SO super simple and easy to model and provides clarity that other models do not. Thank you Ben and everyone that makes TPAW great!

+1

[dogagility]

Just wanted to make my occasional post expressing my gratitude for the TPAW model. SO super simple and easy to model and provides clarity that other models do not. Thank you Ben and everyone that makes TPAW great!

Here, here!!

[Ben Mathew]

Just wanted to make my occasional post expressing my gratitude for the TPAW model. SO super simple and easy to model and provides clarity that other models do not. Thank you Ben and everyone that makes TPAW great!

Thank you for your kind words!

[Ben Mathew]

PERSONAL FINANCE EDUCATION IN COLLEGE

Over the last few years, there has been a push within the economics profession to start teaching personal finance at the college level.

This is huge. Financial education has been very lacking. Is there anywhere I can read more about the "push?"

One of the economists behind this is Professor Annamaria Lusardi who recently moved to Stanford from George Washington and is Director of the Financial Freedom Initiative there. She posts frequently about personal finance education at the college and other levels. She was also interviewed in the Rational Reminder Podcast.

There's a new "Teaching Personal Finance" Conference that she is involved in organizing. The agenda from the first year (2022) gives a sense of some of what's going on in this space at the college level:

[...]

This conference was discussed in an article in the New York Times today:

The Key to Better Math Education? Explaining Money, by Peter Coy (Gifting this article, so this link isn't behind the paywall.)

From the article:

Harvard has a personal finance course in the economics department that’s taught by John Campbell, a past president of the American Finance Association. “Traditionally personal finance was regarded as a very sort of hands-on skill that you might teach to people who were going to a technical high school,” he told me. “There is, I would say, a modern movement to reconceive of personal finance as a subject with actually a lot more intellectual content.”

Last year Stanford hired Annamaria Lusardi, an expert in financial literacy, from George Washington University. In October she and other faculty members held a conference on teaching personal finance that included presenters from Dartmouth College, Loyola University in New Orleans, Pennsylvania State University, the University of Pennsylvania’s Wharton School and the Yale School of Management.

[...]
At Harvard, Campbell said, about a quarter of his students are the first in their families to attend college. “They want to bring expertise back to their families,” he said. “They’re some of the best students in the course because they’re interested in everything.”

The 2024 Teaching Personal Finance conference was held yesterday. I found Professor Michael Boskin's discussion about his experience starting and teaching Stanford's personal finance course interesting:

From 2:02:10 to 2:24:06

Demand for the personal finance course at Stanford is high. And the impact, at least anecdotally, seems great.

Professor Boskin also talked about the need for financial education in the military (starting at 2:16:57).

[Admiral Fun]

Is there a way to input a future windfall into the tool?

e.g. at age 60 add $100,000 to the portfolio.

[dorster]

Is there a way to input a future windfall into the tool?

e.g. at age 60 add $100,000 to the portfolio.

Just do it as a $100K income stream that lasts 1 month starting at age 60.

[Admiral Fun]

Is there a way to input a future windfall into the tool?

e.g. at age 60 add $100,000 to the portfolio.

Just do it as a $100K income stream that lasts 1 month starting at age 60.

Can’t add saving after the retirement date.

I can add an income stream but it doesn’t seem to get added to the portfolio. Unless I’m misunderstanding the output?

[Ben Mathew]

Is there a way to input a future windfall into the tool?

e.g. at age 60 add $100,000 to the portfolio.

Just do it as a $100K income stream that lasts 1 month starting at age 60.

Can’t add saving after the retirement date.

I can add an income stream but it doesn’t seem to get added to the portfolio. Unless I’m misunderstanding the output?

It will get added to the portfolio when the windfall date arrives. Check the portfolio balance graph (you can switch graphs using the drop drown next to the graph title). You should see a bump in the portfolio balance on the date the windfall is scheduled.

[Admiral Fun]

It will get added to the portfolio when the windfall date arrives. Check the portfolio balance graph (you can switch graphs using the drop drown next to the graph title). You should see a bump in the portfolio balance on the date the windfall is scheduled.

It worked, thanks!

[buckeye7983]

Hi Ben-

Thank you so much for TPAW. It is a wonderful tool. Please also tell Jacob that it looks amazing and is remarkably fast.

I have a question. Under Income During Retirement section, I have entered a TIPS ladder (and excluded it from Portfolio Balance). When I look at Balance Sheet, the Net Present Value for the ladder is shown. Does this represent what the cost to construct such a ladder would be today? When I compare to the bid/ask price of the same ladder at tipsladder.com, there is a small but meaningful difference between the two numbers.

Thanks again!

[Ben Mathew]

Hi Ben-

Thank you so much for TPAW. It is a wonderful tool. Please also tell Jacob that it looks amazing and is remarkably fast.

Thank you. I will pass that on to Jacob!

I have a question. Under Income During Retirement section, I have entered a TIPS ladder (and excluded it from Portfolio Balance). When I look at Balance Sheet, the Net Present Value for the ladder is shown. Does this represent what the cost to construct such a ladder would be today? When I compare to the bid/ask price of the same ladder at tipsladder.com, there is a small but meaningful difference between the two numbers.

Thanks again!

Yes, the present value of the TIPS ladder in the balance sheet is the model's estimate of the cost of the ladder today. The difference between this estimate and the more precise estimate from tipsladder.com is that the planner assumes a flat yield curve equal to the expected bond return whereas tipsladder.com looks at the full yield curve.

At the planner's default setting, the expected bond return assumption is the 20 year TIPS yield which is currently 1.8% real. So the value of the ladder shown in the balance sheet is what the ladder would cost if the yield curve were flat at 1.8%. But the current 5 year yield is 1.5% and the 30 year yield is 2.0%. So the actual cost can be higher or lower than the planner estimate depending on the timing of the ladder.

[Raspberry-503]

Remember that you're taking a guess at planning the next 20 or 50 years, there can be an illusion of precision when using there tools.

[stuper1]

Apologies for asking a question that must have been asked several times before. I tried doing a site search to find previous answers, but I didn't turn anything up. My question is: can TPAW be used conveniently to model Roth conversions?

[Ben Mathew]

Apologies for asking a question that must have been asked several times before. I tried doing a site search to find previous answers, but I didn't turn anything up. My question is: can TPAW be used conveniently to model Roth conversions?

Roth conversions cannot be modeled yet because we haven't incorporated taxes yet. Taxes is on the list after duration matching. You will be able to model Roth conversions after that's implemented.

[stuper1]

Apologies for asking a question that must have been asked several times before. I tried doing a site search to find previous answers, but I didn't turn anything up. My question is: can TPAW be used conveniently to model Roth conversions?

Roth conversions cannot be modeled yet because we haven't incorporated taxes yet. Taxes is on the list after duration matching. You will be able to model Roth conversions after that's implemented.

Wonderful to hear. Sorry for all the questions, but do you have a ballpark idea on the timeframe when taxes will be incorporated? Are we talking next year, next 3 years, next 5 years? Thanks for all you guys do.

[Ben Mathew]

Apologies for asking a question that must have been asked several times before. I tried doing a site search to find previous answers, but I didn't turn anything up. My question is: can TPAW be used conveniently to model Roth conversions?

Roth conversions cannot be modeled yet because we haven't incorporated taxes yet. Taxes is on the list after duration matching. You will be able to model Roth conversions after that's implemented.

Wonderful to hear. Sorry for all the questions, but do you have a ballpark idea on the timeframe when taxes will be incorporated? Are we talking next year, next 3 years, next 5 years? Thanks for all you guys do.

No problem. My guess is we can get duration matching done around the end of this year. So maybe taxes by the middle of next year.

[buckeye7983]

Hi Ben-

Thank you so much for TPAW. It is a wonderful tool. Please also tell Jacob that it looks amazing and is remarkably fast.

Thank you. I will pass that on to Jacob!

I have a question. Under Income During Retirement section, I have entered a TIPS ladder (and excluded it from Portfolio Balance). When I look at Balance Sheet, the Net Present Value for the ladder is shown. Does this represent what the cost to construct such a ladder would be today? When I compare to the bid/ask price of the same ladder at tipsladder.com, there is a small but meaningful difference between the two numbers.

Thanks again!

Yes, the present value of the TIPS ladder in the balance sheet is the model's estimate of the cost of the ladder today. The difference between this estimate and the more precise estimate from tipsladder.com is that the planner assumes a flat yield curve equal to the expected bond return whereas tipsladder.com looks at the full yield curve.

At the planner's default setting, the expected bond return assumption is the 20 year TIPS yield which is currently 1.8% real. So the value of the ladder shown in the balance sheet is what the ladder would cost if the yield curve were flat at 1.8%. But the current 5 year yield is 1.5% and the 30 year yield is 2.0%. So the actual cost can be higher or lower than the planner estimate depending on the timing of the ladder.

Thanks for the quick and thorough response, Ben!

[Badgered]

While playing around with this model I notice that the recommended asset allocation changes over time.
Is there a way to model a static asset allocation? For instance, if we are using a balanced one-fund portfolio where the asset allocation stays close to a target, say 60/40 for VSMGX?

(sorry if this has been answered previously ... it's a long thread lol)

[Fireishere]

While playing around with this model I notice that the recommended asset allocation changes over time.
Is there a way to model a static asset allocation? For instance, if we are using a balanced one-fund portfolio where the asset allocation stays close to a target, say 60/40 for VSMGX?

(sorry if this has been answered previously ... it's a long thread lol)

Go to ”advanced” -> ”strategy” -> choose ”SPAW”

[Badgered]

While playing around with this model I notice that the recommended asset allocation changes over time.
Is there a way to model a static asset allocation? For instance, if we are using a balanced one-fund portfolio where the asset allocation stays close to a target, say 60/40 for VSMGX?

(sorry if this has been answered previously ... it's a long thread lol)

Go to ”advanced” -> ”strategy” -> choose ”SPAW”

Thanks

[Raspberry-503]

From what I read the Monte Carlo simulation is done by drawing segments of historical data strung together.
Does that apply to the expected rate of return? For example if I choose 1/CAPE for stocks, does the simulation know the CAPE for each month of historical data and recalculates the AA accordingly? It seems that the discontinuities when jumping to a new samples would create jumps the CAPE Ratio would never experience (whereas the discontinuities don't matter for stock returns themselves if they are a pseudo-random walk)
I assume that similarly inflation is pulled from the same historical samples as the returns?

Edit: thinking about it, you can't use the (supposed) predictive power of 1/CAPE using "random" data, or it wouldn't be random.if if directionally predicted. I'm guessing the Monte Carlo analysis uses today's return over the simulation. Change it or run it again next year to get a new Monte Carlo analysis?
I assume inflation *is* sampled?