How Jon Guyton's firm does drawdown (Guyton-Klinger rules JG)

jamesd

While better known for the Guyton-Klinger rules, Jon (Jonathan) Guyton also has a financial advice practice. Here's their usual plan for how they split a pot at retirement:

1. Safe investments or cash for things like covering the time until state pension age or deferral. Social Security in their case, which can be taken early but rewards waiting.

2. Strict application of the Guyton-Klinger rules for routine spending, including routine holidays and luxuries, not just essentials. They don't routinely add the inflation increases and few people ask for them, which helps with 3.

3. The rest is a discretionary spending lump sum to be used whenever and however the retiree wants to use it, with the understanding that there's no more once it's gone. Asking for more of this happens and either using the gain of no inflation increases, reducing routine future spending or gains from better than average market performance can sometimes fund this.

He explains his firm's approach in the podcast discussion with Michael Kitces at #FASuccess Ep 109: How A Retirement Researcher Implements Retirement Planning For His Own Clients, With Jon Guyton .

This post was prompted by the dilemma expressed in  the topic Are we mad? to which the answer is no if Guyton's approach is followed.

jamesd

The podcast mentions a Withdrawal Policy Statement. That's described in more detail in Guyton's paper The Withdrawal Policy Statement (direct link) on his firm's web site.

Some details of implementing the Guyton-Klinger rules are covered in it.

michaels

For some of us the pre state pension period (1) can be 20 years, no inflation protection for this whole period could prove costly.  I am thinking best buy savings accounts or very short gilts (bills) may be the best way to address this issue, index linked gov bonds are definitely a very expensive way to buy this protection.

ErinGoBrath

michaels said:

For some of us the pre state pension period can be 20 years, no inflation protection for this whole period could prove costly.  I am thinking best buy savings accounts or very short gilts (bills) may be the best way to address this issue, index linked gov bonds are definitely a very expensive way to buy this protection.

5 years in cash and the rest in equities would be better. Gilts are overpriced at the moment.

Thrugelmir

michaels said:

  I am thinking best buy savings accounts or very short gilts (bills) may be the best way to address this issue, index linked gov bonds are definitely a very expensive way to buy this protection.

Treasury Bills are in effect US Gilts. If the fund isn't hedged then currency movement will be a major downside risk. Index linked gilts may prove beneficial in the short term as Central Banks allow inflation some headroom. Equities are hardly a giveaway currently. With expectations of recovery already baked in. 

NedS

jamesd said:

While better known for the Guyton-Klinger rules, Jon (Jonathan) Guyton also has a financial advice practice. Here's their usual plan for how they split a pot at retirement:

1. Safe investments or cash for things like covering the time until state pension age or deferral. Social Security in their case, which can be taken early but rewards waiting.

2. Strict application of the Guyton-Klinger rules for routine spending, including routine holidays and luxuries, not just essentials. They don't routinely add the inflation increases and few people ask for them, which helps with 3.

3. The rest is a discretionary spending lump sum to be used whenever and however the retiree wants to use it, with the understanding that there's no more once it's gone. Asking for more of this happens and either using the gain of no inflation increases, reducing routine future spending or gains from better than average market performance can sometimes fund this.

He explains his firm's approach in the podcast discussion with Michael Kitces at #FASuccess Ep 109: How A Retirement Researcher Implements Retirement Planning For His Own Clients, With Jon Guyton .

This post was prompted by the dilemma expressed in  the topic Are we mad? to which the answer is no if Guyton's approach is followed.

The main issue I see with Guyton-Klinger is that massive flexibility is required as the withdraw rates are based on current portfolio valuations +/- guard rails. So if my portfolio drops 50%, and I also hit the guard rail with dictates I reduce spending by a further 10%, my withdraw amount for that year has just dropped by 60%. That may be fine if I have the flexibility in my spending to accommodate a 60% drop in income (and I guess is the reason I can start with a higher initial withdraw rate and never run out of cash), but I would imagine many don't. I guess if you start off with a higher withdraw rate, you will fundamentally have to make greater compromises down the line if the markets are against you. Unless I've fundamentally misunderstood how Guyton-Klinger works?

I wonder how effective less drastic corrective measures are with a conventional 4% plus CPI drawdown strategy, such as foregoing CPI inflation increases in years with a negative return, or maybe taking a 10% cut in income in very bad years, something which may be far more manageable for most people.

jamesd

michaels said:

For some of us the pre state pension period (1) can be 20 years, no inflation protection for this whole period could prove costly.  I am thinking best buy savings accounts or very short gilts (bills) may be the best way to address this issue, index linked gov bonds are definitely a very expensive way to buy this protection.

The most relevant part of the transcript is:

"maybe you’re retired, one spouse is 67, the other is 66. The 66-year-old has started claiming Social Security, the 67-year-old were waiting until you are 70 to claim, but you need that income, and so you have, we call it the bridge portfolio. And this is a pot of money that’s designed to give you exactly what that defined benefit plan will give you until it does. And then at the time that all of those checks are coming in from those other sources, that fund has run itself out of money by design.

Michael: Interesting. So I can see why you end out with literally, like, three different portfolios with the three different reports, because they just have substantially different time horizons. Like, your bridge portfolio is a classic, you know, asset/liability matched, you know, defined payments portfolio of generally a short term"

For twenty years you might substitute a 100% success rate 4% rule portfolio with a few years of cash in the bond portion that provides short term certainty, though Guyton isn't a fan of holding much cash.

jamesd

NedS said:

The main issue I see with Guyton-Klinger is that massive flexibility is required as the withdraw rates are based on current portfolio valuations +/- guard rails. So if my portfolio drops 50%, and I also hit the guard rail with dictates I reduce spending by a further 10%, my withdraw amount for that year has just dropped by 60%. That may be fine if I have the flexibility in my spending to accommodate a 60% drop in income (and I guess is the reason I can start with a higher initial withdraw rate and never run out of cash), but I would imagine many don't. I guess if you start off with a higher withdraw rate, you will fundamentally have to make greater compromises down the line if the markets are against you. Unless I've fundamentally misunderstood how Guyton-Klinger works?

I wonder how effective less drastic corrective measures are with a conventional 4% plus CPI drawdown strategy, such as foregoing CPI inflation increases in years with a negative return, or maybe taking a 10% cut in income in very bad years, something which may be far more manageable for most people.

That first paragraph is not remotely close to how Guyton-Klinger works. It works more like the second paragraph.

For the first year only you use a percentage of the portfolio value to calculate income. Subsequent years use last year in Pounds to calculate what the new income in Pounds will be after inflation.

You work out what that Pound value is as a percentage of the new portfolio value. If the starting withdrawal rate was 5% then the upper guard rail is 6% of the current portfolio value. If this year's calculated spending is more than six percent then you reduce that planned spend to 90% of its value. As long as it remains above six percent you take that ten percent cut each year.

Say you have 100,000 and start out with a 5% rate, upper guardrail is 6%. Initial income is £5,000. Say inflation is 5% and the market drops 30%. £5,000 + 5% inflation is £5,250. Current portfolio value is £66,500 (70% of the £95,000 left after the first year's £5,000). £5,250 is 7.9% of £66,500 so the upper guard rail is exceeded and the actual next year spending is cut by ten percent to £4,725.

That £4,725 is the base income level used for the calculation in the following year. Let's say same inflation and no market recovery. Planned income is £4,961. Portfolio value is now £4,725 less, so £61,775. £4,961of that is 8%. This is more than 6% so the planned income is cut by 10% to £4,464. 

These cuts continue until recovery or cuts get the income below the 6% upper guard rail.

There are variants that cut by more than ten percent. Those more rapidly get below the guard rail and can reduce the eventual depth of the cut if there's no recovery.

NedS

NedS said:

jamesd said:

While better known for the Guyton-Klinger rules, Jon (Jonathan) Guyton also has a financial advice practice. Here's their usual plan for how they split a pot at retirement:

1. Safe investments or cash for things like covering the time until state pension age or deferral. Social Security in their case, which can be taken early but rewards waiting.

2. Strict application of the Guyton-Klinger rules for routine spending, including routine holidays and luxuries, not just essentials. They don't routinely add the inflation increases and few people ask for them, which helps with 3.

3. The rest is a discretionary spending lump sum to be used whenever and however the retiree wants to use it, with the understanding that there's no more once it's gone. Asking for more of this happens and either using the gain of no inflation increases, reducing routine future spending or gains from better than average market performance can sometimes fund this.

He explains his firm's approach in the podcast discussion with Michael Kitces at #FASuccess Ep 109: How A Retirement Researcher Implements Retirement Planning For His Own Clients, With Jon Guyton .

This post was prompted by the dilemma expressed in  the topic Are we mad? to which the answer is no if Guyton's approach is followed.

The main issue I see with Guyton-Klinger is that massive flexibility is required as the withdraw rates are based on current portfolio valuations +/- guard rails. So if my portfolio drops 50%, and I also hit the guard rail with dictates I reduce spending by a further 10%, my withdraw amount for that year has just dropped by 60%. That may be fine if I have the flexibility in my spending to accommodate a 60% drop in income (and I guess is the reason I can start with a higher initial withdraw rate and never run out of cash), but I would imagine many don't. I guess if you start off with a higher withdraw rate, you will fundamentally have to make greater compromises down the line if the markets are against you. Unless I've fundamentally misunderstood how Guyton-Klinger works?

Quoting myself, it appears I have perhaps misunderstood. I read the following:

https://earlyretirementnow.com/2017/02/08/the-ultimate-guide-to-safe-withdrawal-rates-part-9-guyton-klinger/

which appears to advocate constantly recalculating the rate as current withdrawal amount divided by current portfolio value (rather than basing on last years withdraw amounts) which doesn't appear to be what Guyton-Klinger rules suggest.

jamesd

And while right In some respects, that blog makes some serious mistakes in describing Guyton-Klinger and their choice of approximations and skipping rules is why the thing they wrongly describe as Guyton-Klinger does badly. It comes up often enough that I wrote a post describing some of the issues.

Thrugelmir

jamesd said:

And while right In some respects, that blog makes some serious mistakes in describing Guyton-Klinger and their choice of approximations and skipping rules is why the thing they wrongly describe as Guyton-Klinger does badly. It comes up often enough that I wrote a post describing some of the issues.

But they crunched hard data and experimented, comparing the outcomes. Did you when articulating a response ?