Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
bobcat2
Posts: 4956
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by bobcat2 » Mon Nov 13, 2017 9:38 am

cherijoh wrote:
Mon Nov 06, 2017 6:10 pm
bobcat2 wrote:
Mon Nov 06, 2017 12:59 pm
cherijoh wrote:
Sun Nov 05, 2017 2:56 pm
How do you work with funded ratio when you have income streams coming on at different times?

My current plan is to delay SS until age 70 which will be over 10 years after I plan to retire. Should I figure out a dollar amount to bridge the gap...
Here's what I suggest doing for those delaying SS during early retirement.

1.) For the retirement income coming from the portfolio when you are also receiving SS. -
Get the price of a delayed real life annuity for that stream of retirement income beginning at retirement.

2.) For the extra income stream withdrawn from the portfolio from the retirement date to the time SS benefits begin, estimate the PV of that limited time income stream.

Sum the PV of the limited time income stream with the price of the annuity. That sum is the denominator of the funded ratio.
thanks Bob. Do you have a recommendation for the rate in the PV calculation?
The yield on TIPS whose duration is equal to the duration of the limited time income stream.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

cherijoh
Posts: 4065
Joined: Tue Feb 20, 2007 4:49 pm
Location: Charlotte NC

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by cherijoh » Mon Nov 13, 2017 11:56 am

bobcat2 wrote:
Mon Nov 13, 2017 9:38 am
cherijoh wrote:
Mon Nov 06, 2017 6:10 pm
thanks Bob. Do you have a recommendation for the rate in the PV calculation?

The yield on TIPS whose duration is equal to the duration of the limited time income stream.

BobK
Thanks Bob!

User avatar
Portfolio7
Posts: 265
Joined: Tue Aug 02, 2016 3:53 am

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by Portfolio7 » Tue Nov 14, 2017 11:12 am

Hi Bob, thanks for this great piece! I ran an initial calculation, just to see what it looked like, and now I'm trying to make sure I'm not making stupid mistakes in my calculations.

I'm 50 years old and hope I'm roughly 12 years from retirement. I calculated the percentage of current savings vs the cost of a deferred life annuity with 3% cola, joint with DW, to start in 12 years. That yields 32%. Does that seem a reasonable FR result when we have about 1/3 of our target savings amount (which I understand that one point of FR is to move away from using a target number, but I am trying to provide a base of understanding)? If I calc savings vs our 'target number', I'm about 35% funded, as a comparison, and I would think it makes sense that the two numbers are similar? ... and that makes me think the FR Calculation I'm doing may be about right? Does 32% sound reasonable for someone who is 10-15 years from retirement or am I way off track on the calculation? In terms of what the 32% means, I think it means we need to step up the saving slightly if we want to hit the retirement income goal that we set in the timeline we hoped for. I'm not sure if there is some kind of FR % guide as to where you should be given X years from retirement?
An investment in knowledge pays the best interest.

User avatar
siamond
Posts: 3469
Joined: Mon May 28, 2012 5:50 am

Re: Using the Funded Ratio to drive your retirement investment plan

Post by siamond » Tue Nov 14, 2017 11:49 pm

bobcat2 wrote:
Sun May 28, 2017 9:52 am
The funded ratio (FR) is a metric that can be used to monitor your retirement investment plan. It is simply the ratio of assets to liabilities, where the assets are the size of your portfolio and the liabilities are the present value of your targeted retirement income stream.
I never went very far in those threads because it quickly goes into TIPS and Annuities and Liability Matching and what not, concepts that just turn me off real fast. Now it struck me that maybe we should step back and focus on this core definition of the funded ratio, in the context of nisiprius' interesting comment (for which I didn't see a satisfying answer while parsing this thread).
nisiprius wrote:
Mon May 29, 2017 10:06 am
On the other hand, just to point out the obvious: how do you "use the Funded Ratio to drive your retirement investment plan" when the funded ratio is exactly as volatile as your portfolio?
Let me first suggest to refine the denominator of the funded ratio as the present value of the cash flows envisioned during retirement. Annual desired income should be counted as positive values (i.e. liabilities). Extra spend for LTC years (if self-insuring) should be counted as positive values (more liabilities). Fixed income like SSA/Pension (and yes, annuities!), when they start, should be counted as negative values (this reduces liabilities). Occasional lump sums like an inheritance should be counted as negative values. Extending the logic, if one spouse is early retired, and the other still works, then the wages of the working spouse should be counted as negative values. In short, model ALL future cash flows, and compute the net present value (NPV).

Then we can compute the funded ratio by dividing the portfolio current balance by the NPV math. And yes, big question is... is it more or less than 1? But I brushed over one crucial detail. What discount rate was used for the NPV math? 4%? 5%? How was this determined? Instead of deflecting the answer to the annuity industry, why not call a cat a cat? This is the expected return of the portfolio during the retirement period being modeled, simple as that. Which is directly dependent on the asset allocation, and ALSO directly dependent on how over/undervalued the market is.

Say we have a given portfolio balance of $2M. Considering the AA (say 60/40), we use a 4% discount/expected rate (being suspicious about an ongoing long bull market) on the future cash flows, and the funded ratio is more than 1, cool. Then the market crashes, and said portfolio is down to $1.4M (ouch!). Well, the expected returns are HIGHLY dependent on the starting point, as history demonstrates. The funded ratio didn't really change in truth. What changed was the expected returns (much higher!). And nisiprius' observation is now answered.

Bottomline: if one computes the funding ratio based on future cash flows, by using a fairly solid way of assessing expected returns (say 1/CAPE for stocks, and inflation-adjusted yield for bonds) to get to a discount rate, then the funded ratio core concept seems sound (and useful), and is not really subject to the portfolio's volatility. This being said with the usual caveats about expected returns models, and yes, this a tough subject in itself...

I'll just finish by suggesting that the funded ratio is basically the converse of actuarial withdrawal methods (PMT-based, like VPW), and that such methods are quite robust to an inaccurate guess of expected returns. And that none of such logic requires annuities or TIPS or what not to be useful for retirees and pre-retirees assessing their portfolio vs. desired income streams.

Peter G
Posts: 42
Joined: Sat Apr 30, 2016 8:14 pm

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by Peter G » Wed Nov 15, 2017 6:18 am

nisiprius wrote: ↑Tue May 30, 2017 1:06 am
On the other hand, just to point out the obvious: how do you "use the Funded Ratio to drive your retirement investment plan" when the funded ratio is exactly as volatile as your portfolio?
If you're at/near/in retirement the options might be:
1. convert all assets into inflation matching, low risk liability matching portfolio of annuity, tips etc.
2. use the approach of Lussier in Successful investing is a Process, which favours a LMP approach, but says it has low expected returns and thus prefers some assets in LMP and some in an at-risk portfolio (the mix depending on risk aversion and how well covered the liabilities are). He proposes a dynamic management of all the assets, moving from the at-risk to the LMP depending on value measures like PE10 and yield. He also likes to track asset class volatility and correlations and make AA adjustments in the at-risk portfolio accordingly. That seems a bit complicated for me.
3. In Retirement Portfolios: Theory, Construction and Management, Michael Zwecher suggests the bigger the FR, the less important it is to secure the safe asset spending 'floor'. He also allows putting some assets into big risk/big return if the floor is very inadequate, but selling them if they fall so as to threaten a yet lower basic living floor. You can lose out by having sold low but that’s the risk you took to try to get a higher floor when your savings were inadequate.
I hope I haven't misrepresented them.
I imagine that many folk who believe in the income safety inherent in liability driven investing wouldn't be that keen on using measures like 1/CAPE, or the volatility of stock prices for funding estimates.

User avatar
bobcat2
Posts: 4956
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by bobcat2 » Fri Nov 17, 2017 10:52 am

Portfolio7 wrote:
Tue Nov 14, 2017 11:12 am
I'm 50 years old and hope I'm roughly 12 years from retirement. I calculated the percentage of current savings vs the cost of a deferred life annuity with 3% cola, joint with DW, to start in 12 years. That yields 32%. Does that seem a reasonable FR result when we have about 1/3 of our target savings amount (which I understand that one point of FR is to move away from using a target number, but I am trying to provide a base of understanding)? If I calc savings vs our 'target number', I'm about 35% funded, as a comparison, and I would think it makes sense that the two numbers are similar? ... and that makes me think the FR Calculation I'm doing may be about right? Does 32% sound reasonable for someone who is 10-15 years from retirement or am I way off track on the calculation? In terms of what the 32% means, I think it means we need to step up the saving slightly if we want to hit the retirement income goal that we set in the timeline we hoped for. I'm not sure if there is some kind of FR % guide as to where you should be given X years from retirement?
Calculating the funded ratio when a couple is involved is more complicated than for a single person. When two people live together their individual living standards are higher than if they were living separately. In economics it's called 'the economies of shared living’. This means that when keeping the living standard constant for the surviving partner living alone, he or she will typically need 65% - 75% of the income the couple had, not 100% or 50%, to keep the same living standard.

I don’t know how the joint life annuity you priced handles the survivor’s income benefit, but typically joint life annuities do not drop the level of the income benefit when the first partner dies. This makes such annuities overprice retirement income for couples. The annuity is providing 100% of the couple’s income for the lone survivor, not 65% - 75%.

Here’s a method for calculating the funded ratio for couples. Estimate the income goal for the couple and also estimate how much income the survivor would need to keep the same living standard. I suggest a default value of 75% of the couple’s income for the survivor’s income goal. Price a joint life annuity that keeps the income benefit constant once the first partner dies for the 75% income goal. For the other 25% price a single life annuity for one partner, or calculate the average price of a single life for each of the partners. The price of the sum of the two life annuities (joint & single) is the denominator of the funded ratio for couples.
Does 32% sound reasonable for someone who is 10-15 years from retirement or am I way off track on the calculation?
It seems low to me, but some of the problem may be that the joint life annuity you priced is overstating the amount of income you need, which is the point I addressed above.

I'm not sure if there is some kind of FR % guide as to where you should be given X years from retirement?
See this earlier post in the thread about projecting the funded ratio at retirement.
Link - viewtopic.php?f=10&t=219878&p=3613708&h ... k#p3613708

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 4956
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by bobcat2 » Sat Nov 18, 2017 7:12 pm

Calculating Retirement Income: Beware "Ballpark" Calculations
"Best-guess" financial planning using rule-of-thumb guidelines can miss retirement targets by a mile.
For years, advisors have relied on the "4% rule," the rule of thumb that dictates a 4% annual retirement income drawn from investment assets. If only it were that easy in reality. “Common approaches like the ‘4% rule’ are easy to understand, but do not account for a client’s individual circumstances and can lead to unintended mistakes,” says Rod Greenshields, consulting director with Russell Investments. “We think advisors would do well to follow the lead taken by defined benefit plans and calculate a funded ratio. By determining the cost of a client’s liabilities compared to the value of their assets, the funded ratio offers a superior method of evaluating retirement readiness. The math behind the ratio is sophisticated, but the outcome is a simple yet powerful percentage that most clients understand immediately. This individualized approach gives advisors the opportunity to engage with clients in a meaningful conversation about their progress toward retirement goals and the amount of risk they may need to take in order to meet them.”
Link to article - https://www.thestreet.com/story/1295497 ... tions.html

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 4956
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Using the Funded Ratio to drive your retirement investment plan

Post by bobcat2 » Mon Nov 20, 2017 1:42 pm

Now it struck me that maybe we should step back and focus on this core definition of the funded ratio, in the context of nisiprius' interesting comment (for which I didn't see a satisfying answer while parsing this thread).


Nisiprius' comment.
nisiprius wrote:
Mon May 29, 2017 10:06 am
On the other hand, just to point out the obvious: how do you "use the Funded Ratio to drive your retirement investment plan" when the funded ratio is exactly as volatile as your portfolio?
I thought I had refuted this somewhere, but in case I haven't, here goes.

It is simply not true that the funded ratio is exactly as volatile as your portfolio. If interest rates fall the value of the bonds in the portfolio rises, but the cost of the annuity also rises. In other words, the bond price rise in the numerator is more than offset by the rise in the price of the annuity in the denominator. In that case the portfolio is worth more, but the funded ratio falls. Hardly the same volatility.

A similar relationship holds in the converse case. If interest rates rise, the value of the portfolio falls, but the price of the life annuity also falls, and the funded ratio rises. Again the volatility of the portfolio is quite different from that of the funded ratio.

BobK

PS - The above analysis assumes the duration of the bonds is less than the duration of the annuity, which would typically be the case unless one was holding long-term bonds. In the case where the duration of the bonds and the annuity was matched, the price changes of the bond and the annuity would offset and there would be less change in the funded ratio. However, it wouldn't be 100% offset unless you were holding all bonds because the interest rate change affects all of the liabilities in the denominator, but only a portion of the assets in the numerator. In the relative unusual case of holding very long-term bonds whose duration was longer than that of the annuity, the above analysis would be reversed. How the funded ratio would be affected, however, would also depend on the ratio of bonds to stocks in the portfolio.

Edited for PS.
Last edited by bobcat2 on Tue Nov 21, 2017 9:44 am, edited 2 times in total.
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
siamond
Posts: 3469
Joined: Mon May 28, 2012 5:50 am

Re: Using the Funded Ratio to drive your retirement investment plan

Post by siamond » Mon Nov 20, 2017 3:04 pm

bobcat2 wrote:
Mon Nov 20, 2017 1:42 pm
siamond wrote: Now it struck me that maybe we should step back and focus on this core definition of the funded ratio, in the context of nisiprius' interesting comment (for which I didn't see a satisfying answer while parsing this thread).

Nisiprius' comment.
nisiprius wrote:
Mon May 29, 2017 10:06 am
On the other hand, just to point out the obvious: how do you "use the Funded Ratio to drive your retirement investment plan" when the funded ratio is exactly as volatile as your portfolio?
I thought I had refuted this somewhere, but in case I haven't, here goes.

It is simply not true that the funded ratio is exactly as volatile as your portfolio. If interest rates fall the value of the bonds in the portfolio rises, but the cost of the annuity also rises. In other words, the bond price rise in the numerator is more than offset by the rise in the price of the annuity in the denominator. In that case the portfolio is worth more, but the funded ratio falls. Hardly the same volatility.

A similar relationship holds in the converse case. If interest rates rise, the value of the portfolio falls, but the price of the life annuity also falls, and the funded ratio rises. Again the volatility of the portfolio is quite different from that of the funded ratio.
Ok, maybe the 'exactly' qualifier wasn't entirely exact. But the point remains. When the portfolio value crashes due to a 30% or 50% (or more) equity crash, interest rates may move or may not, but the end result on the 'funded ratio' (as you defined it) will typically be VERY significant because of the dramatic change in the portfolio's value. Maybe this is actually fair, because it puts a retirement strategy centered on annuities on TIPS on a more difficult footing. Still, it is a little difficult to accept that an advanced indicator of retirement readiness would go up & down depending on stock valuations.

Now, if we come to a core definition like I did in my post, then things become mathematically consistent. Maybe I should call this a "Present Value Ratio" to clearly distinguish from the concept you're promoting. I actually played with such "PV Ratio" metric in the past few days, and I find it extremely handy, for retirees or future retirees not eager to go down the LMP route. So... even if this clearly strongly departs from your recommendations, thank you for the idea!

EDIT: fixed the link.
Last edited by siamond on Mon Nov 20, 2017 3:28 pm, edited 1 time in total.

sandramjet
Posts: 167
Joined: Thu Oct 23, 2014 11:28 pm

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by sandramjet » Mon Nov 20, 2017 3:22 pm

siamond wrote:
Mon Nov 20, 2017 3:04 pm
Now, if we come to a core definition like I did in my post, then things become mathematically consistent.
What post is that? The link seems to go to just a reply form...

User avatar
siamond
Posts: 3469
Joined: Mon May 28, 2012 5:50 am

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by siamond » Mon Nov 20, 2017 3:30 pm

sandramjet wrote:
Mon Nov 20, 2017 3:22 pm
siamond wrote:
Mon Nov 20, 2017 3:04 pm
Now, if we come to a core definition like I did in my post, then things become mathematically consistent.
What post is that? The link seems to go to just a reply form...
Oops, sorry, my bad. I fixed the faulty post. Here is the right link.

User avatar
bobcat2
Posts: 4956
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by bobcat2 » Mon Nov 20, 2017 3:47 pm

The funded ratio tells you what percentage of your retirement income goal you can purchase today. If you are at or near 100% your retirement income goal,why would you want to be heavily invested in stocks? The funded ratio is telling you to lighten up when you are close to your income goal, particularly if you are also close to your targeted retirement date.
Still, it is a little difficult to accept that an advanced indicator of retirement readiness would go up & down depending on stock valuations.
Whether it is difficult for you to accept or not, it's true that big swings in stock valuations, if your portfolio is heavily weighted in stocks and you are within 12 years or less to your retirement date, will have a big effect on retirement readiness.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
siamond
Posts: 3469
Joined: Mon May 28, 2012 5:50 am

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by siamond » Mon Nov 20, 2017 3:56 pm

bobcat2 wrote:
Mon Nov 20, 2017 3:47 pm
Whether it is difficult for you to accept or not, it's true that big swings in stock valuations, if your portfolio is heavily weighted in stocks and you are within 15 years or less to your retirement date, will have a big effect on retirement readiness.
I actually strongly disagree. I think our disconnect is simple, you're thinking in an LMP context and I'm thinking in a fixed AA context (for life, which doesn't stop the day you retire). I understand your view, just saying that one can do a Portfolio/PV calculation that is mathematically consistent, and that applies to other retirement strategies.

User avatar
bobcat2
Posts: 4956
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by bobcat2 » Mon Nov 20, 2017 4:55 pm

Why the expected return on the portfolio should not be used as the discount rate for the liabilities in the denominator of the funded ratio.

From a paper by Biggs & Smetters on why S&L pensions should not do this.
In fact, most economists believe that pension liabilities should be valued using low discount rates even if pensions continue to invest in stocks and other risky but high-return assets and even if their assumptions regarding future average investment returns are accurate. However, the theories and assumptions underlying a fair-market valuation approach are so ingrained among economists that, in many cases, the proponents fail to make these background arguments explicit. …


Put another way, GASB accounting rules help pension plans calculate a “best guess” annual contribution that gives the plan a roughly 50 percent probability of being able to meet its benefit obligations. Assuming the plan has correctly pegged the ultimate return on its investments, over any given period of time—even decades—the annual fluctuations in returns mean that the return actually received is almost sure to be either above or below the projection. Thus, even a “fully funded” plan has only a roughly 50–50 chance of generating returns sufficient to pay full benefits. But
legally guaranteed pension benefits must be paid with 100 percent certainty. …

Economists argue that the discount rate used to value future pension liabilities should reflect the fact that pension benefits are guaranteed, even if the returns on a pension’s investments are not. More formally, the discount rate applied to the liability should be based on the risk of the liability, not the risk of any assets used to fund that liability. While there is some disagreement regarding how risky accrued public pension benefits actually are, economists are united in believing that the appropriate discount rate is a function of the liability rather than of the assets.

Moreover, this is how financial markets value liabilities. If a pension sought to transfer its liabilities to a private insurance company— something common in the United Kingdom, though for tax reasons not in the United States—the insurer would base its price on the size and risk of the liabilities, without reference to the risk or expected return of the asset backing those liabilities. The reason is that the investment portfolio can be changed at any time to any combination of risk and return the pension chooses, but the liabilities, if guaranteed, must be paid regardless. ...

The fair-market valuation method tells us the costs of achieving full funding. And it incorporates a number of truths from economics and finance: that stocks pay higher returns because they are more risky; that the risk of stock investments does not disappear over long holding periods; and that individuals, either as investors or taxpayers, do not value gains equally with losses. Finally, valuing guaranteed pension liabilities using a riskless discount rate does not imply that pensions must invest only in bonds. Rather, it merely shows that the value of a guaranteed pension benefit is independent of the returns on risky investments used to fund that benefit.
Link - https://www.aei.org/wp-content/uploads/ ... 782445.pdf

In short, it is irrelevant whether you can invest in the safest asset or not in finding the PV of the liability. If the liability is meant to be met with high probability then the return of the low risk asset is the proper discount rate. The expected returns of the portfolio is a separate issue. If that weren't the case, then if my 50/50 portfolio gave me a below 100% funding ratio I could solve my problem to shifting to a 100/0 portfolio where the 50% bonds are replaced by the highest expected return asset class, emerging market stocks. Funding ratio problem solved. :D :wink:

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 4956
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by bobcat2 » Mon Nov 20, 2017 4:56 pm

Why the expected return on the portfolio should not be used as the discount rate for the liabilities in the denominator of the funded ratio. cont.

From an article by financial planner Paula Hogan.
What is the relevance of these ideas for your own personal financial planning? Or more to the point, what ideas from the pension world do we NOT want to bring over to personal financial planning? First, when planning for retirement, set yourself up for clear decision-making. Separate the exercise of determining the cost of safely funding your income in retirement from the exercise of deciding how to invest your portfolio.

When planning for retirement income, use as your base case a plan that does not rely on stock performance for success. Then, look at the range of possibilities implied as you increase risk from that base income projection. As you increase portfolio risk beyond the base case scenario, you are in essence indicating that you are able and willing to accept a reduced future standard of living if necessary in return for having a chance of obtaining a higher standard of living.

Or, perhaps you will decide to increase portfolio risk even though you are unable or unwilling to reduce your future standard of living if investments don't work out as planned. If you take this route, and like a pension fund use current accounting guidelines rather than a fair-market valuation approach, you are in essence relying on the future-taxpayer/white-knight scenario as your ultimate financial safety net in retirement.

Are public pension plans your model for retirement planning?
Link - http://hoganfinancial.com/wealthinking/ ... -planning/

The expected value of a risky portfolio means about half the time you will get less than the expected value. If there are a lot of stocks in the portfolio there is about a 20% chance you will get significantly less than the expected value. If your estimated expected value is too high you are even worse shape.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 4956
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by bobcat2 » Mon Nov 20, 2017 4:59 pm

Why the expected return on the portfolio should not be used as the discount rate for the liabilities in the denominator of the funded ratio. cont.

Robert Novy-Marx is one of the better known financial economists on this board principally because he is one of Larry Swedroe's favorite economists and Larry often cited Novy-Marx's papers.

Here is Robert Novy-Marx on valuing liabilities of pensions. Keep in mind that for most American workers their 401k is how they receive retirement pension-like income in lieu of a pension. First he considers state and local pensions.
While most public sector actuaries and the National Association of State Retirement Administrators support the Governmental Accounting
Standards Board's (GASB's) methodology, academic economists disagree with its underlying principles with near unanimity. The heart of the disagreement lies with the rate used to discount liabilities. GASB discounts projected benefit payments at the expected return on plan assets, because the Board believes this rate “best reflects the employer’s projected sacrifice of resources” (GASB, 2011b). Financial economists prefer a lower discount rate, because accumulated pension promises are viewed as nearly default-free and the basic tenets of financial economics require that cash-flows are discounted at rates reflecting their risks.


Then there is this footnote.
Much of the actuarial and accounting professions actually support the economists’ position. The Financial Accounting Standards Board, the source of generally accepted accounting principles for the U.S. private sector, specifies that corporations calculate pension liabilities by discounting expected future benefit payments using high grade corporate bond yields. This basically treats pension debt in the same manner as firms’ other debts more generally. The Penson Protection Act of 2006 similarly specifies minimum funding requirements for DB pension plans using liability calculations that employ investment grade corporate bond rates. The Pension Benefit Guarantee Corporation bases the variable insurance premiums it charges plans on these same calculations. Finally, common practice in mergers and acquisitions involves calculating pension liabilities using a “settlement rate” that typically reflects the discount rate implicit in the market price of annuities, which are typically even lower than the yields on high grade corporate debt.
The rest of the paper deals with absurdities that can result from discounting retirement income liabilities by the expected return on invested assets.I particularly liked the strategy of burning money which follows.
A plan can, under GASB logic, sometimes improve its funding status by literally burning money. This result holds because the marginal value of a dollar of assets, as measured by GASB, can be negative. By destroying a dollar’s worth of T-bills, or other cash-equivalents, a manager decreases a plan’s assets, but increases the remaining assets’ expected returns. These higher asset returns decrease the plan’s liability, as recognized by GASB, and this decrease can more than offset the loss of (the burned) assets.
:D :D

Here's a link to the paper
http://rnm.simon.rochester.edu/research/LIoGMfVPL.pdf

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
siamond
Posts: 3469
Joined: Mon May 28, 2012 5:50 am

Re: Using Funded Ratio to drive retirement investment plan - Part 1 of Funded Ratio series

Post by siamond » Mon Nov 20, 2017 6:00 pm

bobcat2 wrote:
Mon Nov 20, 2017 4:55 pm
In short, it is irrelevant whether you can invest in the safest asset or not in finding the PV of the liability. If the liability is meant to be met with high probability then the return of the low risk asset is the proper discount rate. The expected returns of the portfolio is a separate issue.
Well, I'm afraid that some of the quotations you provided are provided out of context. Valuing a pension transferring a set of liabilities to another company has nothing do with the perspective of individual investors looking at their portfolio and future cash flows.

We're not trying to value liabilities in a vacuum here. In the context I was discussing, we're computing a ratio between one's portfolio and one's future cash flows (which will include all sorts of positive numbers like SSA, Pension etc; and where the true liabilities is the desired future spending budget, which should have some level of flexibility, we're not machines, we adapt). In such a case, the units used in the denominator have to be consistent with the units used in the numerator (AND with the retirement strategy being planned for).

The true question is, will the portfolio value and its trajectory over time cover the future cash flows. Then of course we're speaking of the portfolio's future returns, I mean, what else? Sure enough, one would be conservative about such expected returns (NOT use an average, use a low percentile of expected outcomes) to base the discounted math on, but this remains very directly dependent on the portfolio's composition. That is, if we keep a fixed AA portfolio during retirement (instead of going to annuities/etc).

I think this is what some of the other quotes you used are saying and here I agree. I did mean "conservative expected returns (as discount rate)", and maybe I wrote too hastily "expected returns" without qualifier in a few cases (yes, the point at the center of a distribution of probabilities would not be good here). Point granted. Still, this is all about your own portfolio, and not about 100% TIPS or anything else. Otherwise, we compute a ratio between apples and oranges.
bobcat2 wrote:
Mon Nov 20, 2017 4:55 pm
If that weren't the case, then if my 50/50 portfolio gave me a below 100% funding ratio I could solve my problem to shifting to a 100/0 portfolio where the 50% bonds are replaced by the highest expected return asset class, emerging market stocks. Funding ratio problem solved. :D :wink:
Well... this actually... demonstrates my point. The expected returns of a single very volatile investment vehicle like emerging markets should be assessed in a very conservative manner in such context, that is for sure. They may be lower than a regular 50/50 portfolio (this is not entirely obvious to me, but quite possibly). Now if you take a portfolio with 100% bonds, then you'd better be ready for lower liabilities! So yes, of course, one's portfolio composition has a very direct impact on one's portfolio future trajectory, hence on retirement readiness.

Ok, frankly, I think we can go in circles for many posts here... Your head is in LMP mode, my head is in fixed AA mode. They are just different approaches, and can't be characterized by the same metric. Sorry I interfered with your thread and side-tracked. You just gave me an interesting idea without intending to (what I called "PV Ratio"), and I thank you for that.

PS. I now understand better your 'funded ratio' in its intended context, and why it varies due to stock valuations, as a direct function of the retirement strategy being proposed. Wasn't my context, it was yours, and this is clearer now.

Post Reply