Cash buffer yields

[urn2bfree]

This year old post I recently ran across is the impetus to this post.

http://www.kitces.com/blog/index.ph...ss-Youre-A-Good-Market-Timer.html&serendipity[csuccess]=moderate

This discusses and links to a paper purportedly showing that keeping a cash buffer of any significant size (to spend during down years instead of making your usual WR from your portfolio) DECREASES success rates.

I have a problem with this as the paper assumes a very low yield on cash buffers.


I get close to 1% (pre tax) and given current bond yields figure that just makes cash a very short term part of my bond portfolio.

What kind of yield do people get on cash accounts here?

 

[H2ODude]

I had subscribed to the low cash AA because it seemed so stagnant; my cash holdings at FIDO were not even worth tracking. However, I got tired of watching the market every day (well, I still sorta do) and find it easier now that I raised about three years of cash for annual needs beyond pension (and am still pre SS). I put it in Amex online savings, yields 0.85%. What I find interesting is that IIRC the daily savings rate exceeds what they offer for a one year CD. Anyway, I know it doesn't keep up with inflation but it's better than FIDO and I can rest easy knowing ups and downs don't affect what I need for next 3-4 years.

 

[urn2bfree]

No bond strategy is likely to keep up with inflation. The paper I linked to does discuss the psychological effect of a cash buffer but suggest it is irrational to the point that it hurts overall success rates. I think that finding is still largely a by product of the inputs-- they set the yield for cash buffer artificially low...well not artificial,as I am sure you COULD put your cash buffer in such a low yielding account. I wonder what the results would look like if the cash yields were closer to what you (0.85) and I (0.97) are getting compared to current 1 month T bill rates (0.01 to 0.02). Obviously this is jut one snap shot of yields at a very low yield time, but the principle of setting such low yields on cash has to disadvantage the buffer approach from the start. I am wondering what sort of yields people really get on their buffers.

 

[pb4uski]

....I get close to 1% (pre tax) and given current bond yields figure that just makes cash a very short term part of my bond portfolio.

What kind of yield do people get on cash accounts here?

Mine is an online savings account yielding 0.8% and the Vanguard ST IG fund that has an SEC yield of 1.54% and a distribution yield of 1.75%.

Between the two I get ~1.4%

 

[H2ODude]

I work with a WR of ~2.5%; hence if 8-9% of investable assets is sitting on the sidelines making only 0.85% I don't consider it a big drain on performance. Instead I look at it as a way to generate confidence and comfort. Mind you this is a relatively recent phenomena for me; when markets are rising it's hard for me to keep this cash just sitting on sidelines. I guess that's the greedy part of me as opposed to the "let's be safe here" me.

 

[NW-Bound]

I currently have 26% in cash. Most of that money would normally be in bond funds. I still have 5% in a diversified bond fund. It lost 1.1% YTD.

Of that cash, the best yield is from I-bonds that I accumulated a few years ago. Combined, they yielded 2.9% in the past 12 months,

Next is a Stable Value fund in my wife's 401k which yielded 1.6% in the last 12 months.

Finally, my Ford Credit Advantage account has a yield of 1.1%.

 

[Animorph]

Cash will almost always be the lowest growth position in a portfolio AA. It can't help but be a drag on portfolio returns. One bad thing this report did, IIRC, was to refill the cash buffer as soon as the market started to recover, not after it had fully recovered. Not the most efficient refill timing, much closer to just maintaining a fixed cash allocation.

Clearly, keeping a significant cash position throughout retirement is a drag on performance compared to historical equity and bond returns. But used as a true buffer, spent completely down during bad market years and not refilled until after a full recovery, it might be worth a bit more. I don't know of any calculators that will use cash or bonds that way.

 

[bo_knows]

Find a retirement calculator that will accept cash inputs, and lets you put the yield of that cash investment, and see for yourself.

Cash will almost always be the lowest growth position in a portfolio AA. It can't help but be a drag on portfolio returns. One bad thing this report did, IIRC, was to refill the cash buffer as soon as the market started to recover, not after it had fully recovered. Not the most efficient refill timing, much closer to just maintaining a fixed cash allocation.

Clearly, keeping a significant cash position throughout retirement is a drag on performance compared to historical equity and bond returns. But used as a true buffer, spent completely down during bad market years and not refilled until after a full recovery, it might be worth a bit more. I don't know of any calculators that will use cash or bonds that way.

So, if you had a cash buffer and modelled that you would spend it down during a -15% or more drop in a given year, when would you say to refill it? When it was back to the level before the -15%, or some arbitrary % above the -15% mark? Trying to figure out how you'd want to model it.

 

[audreyh1]

Cash will almost always be the lowest growth position in a portfolio AA. It can't help but be a drag on portfolio returns. One bad thing this report did, IIRC, was to refill the cash buffer as soon as the market started to recover, not after it had fully recovered. Not the most efficient refill timing, much closer to just maintaining a fixed cash allocation.

Clearly, keeping a significant cash position throughout retirement is a drag on performance compared to historical equity and bond returns. But used as a true buffer, spent completely down during bad market years and not refilled until after a full recovery, it might be worth a bit more. I don't know of any calculators that will use cash or bonds that way.

Yes - I don't think the point is to refill the cash bucket until after stocks have recovered significantly. Otherwise just seems silly.

 

[LOL!]

Yes - I don't think the point is to refill the cash bucket until after stocks have recovered significantly. Otherwise just seems silly.

Aren't you supposed to use the cash (and bonds) to buy equities before they have recovered significantly?

Isn't that call rebalancing? I see lots of folks happy to rebalance out of equities that have done well. I don't see so many folks happy to rebalance in to equities that have done poorly.

 

[urn2bfree]

Aren't you supposed to use the cash (and bonds) to buy equities before they have recovered significantly?

Isn't that call rebalancing? I see lots of folks happy to rebalance out of equities that have done well. I don't see so many folks happy to rebalance in to equities that have done poorly.

If you do not rebalance INTO equities when they are down, it is not rebalancing... and defeats the entire purpose.

Cash is just a very short bond. One can consider it a part of your bond allocation or as its own special bucket. Either way. It seems to me that the academic research needs to treat the cash buffer as any other part of maintaining an allocation. It should also use realistic returns for each portion of the AA. The academic research I have seen so far does neither.
If your total portfolio is stocks65/bonds35/cash5. When you have a big drop in stocks, it therefore throws the AA off. Spending cash both on day to day expenses and to buy depressed assets (if you have enough to do so) as you shift assets to maintain the AA SHOULD (theroretically) have you improving survival and performance.

 

[Katsmeow]

Contrast this with using bonds to simulate an SPIA and drawing from this when stocks are down and not rebalancing (meaning you have more equities over time):

Understanding The True Impact Of Single Premium Immediate Annuities On Retirement Income Sustainability - Kitces | Nerd's Eye View

In my own case, my cash yields virtually nothing. I have about 9% of my portfolio either in bank accounts (very small amount) or in Prime Money Market at Vanguard. This is where we take ongoing withdrawals out of.

I keep another 8.5% of the portfolio in VTSUX - Short Term Investment Grade which I functionally treat as if it was cash.

The reason I have so much in these is that for the next 3 years our portfolio withdrawals will be unusually high (kids in college) and so I wanted to put enough to fund most of those withdrawals in a very short term source. If the market were to go down during these 3 years I would draw from these and not sell equities.

 

[W2R]

My planned AA includes 5.5% cash, and I am getting almost no yield from it.

Oh well. Equities go up and down, bonds yield more sometimes than others, and cash is yielding almost nothing. But you know, my AA was planned for the long term, and I haven't been chasing yield. I am no good at market timing so I just don't try to do it. My portfolio has been growing as a whole, and I am OK with my AA.

 

[Animorph]

So, if you had a cash buffer and modelled that you would spend it down during a -15% or more drop in a given year, when would you say to refill it? When it was back to the level before the -15%, or some arbitrary % above the -15% mark? Trying to figure out how you'd want to model it.

My strategy is to raise cash when my portfolio is above retirement plan expectations, which just uses a fixed growth rate. Where normally many will sell a portion of their portfolio at the start of the year to pay that year's expenses, I'll sell any time my portfolio reaches any future year's starting value. Kind of like triggered rebalancing instead of yearly. If I have cash, I'll spend it down before touching the portfolio (and by not touching, I mean reinvesting dividends and CG's). When I run out of cash it will be month to month portfolio withdrawals. Obviously dependent on your growth rate assumptions, which can be adjusted.

 

[urn2bfree]

My planned AA includes 5.5% cash, and I am getting almost no yield from it.

Oh well. Equities go up and down, bonds yield more sometimes than others, and cash is yielding almost nothing. But you know, my AA was planned for the long term, and I haven't been chasing yield. I am no good at market timing so I just don't try to do it. My portfolio has been growing as a whole, and I am OK with my AA.

Almost no yield and what the real yield can still be significantly different. The academic paper I cited uses 3 month Treasuries to define cash yields...that would be like 0.2 %. I GET 1% on my credit Union cash. Many online banks are paying at least 0.8%. Those are low yields, but the paper's choice of yield seems unreasonably pessimistic to me.

 

[mathjak107]

The problem with cash is it offers no capital gains appreciation potential when stocks fall.

Bonds usually rise and offset stocks volatility.

If you are comfortable with the volatility range of your portfolio then listening to the noise about not owning bonds and selling them for cash will give you a higher range of volatility.

You may have to sell equities to raise cash levels to offset the swings.

That is where most folks drop the ball. They don't realize they will have to cut equity exposure more by holding more cash .

They would cut their gains on the equities side by more than the potential drop on the bonds side.

 

[rayvt]

Almost no yield and what the real yield can still be significantly different. The academic paper I cited uses 3 month Treasuries to define cash yields...that would be like 0.2 %. I GET 1% on my credit Union cash. Many online banks are paying at least 0.8%. Those are low yields, but the paper's choice of yield seems unreasonably pessimistic to me.

No. The reason for the cash bucket is 100% safety and liquidity, not to make a profit. Therefore it should be in something like short-term Treasuries.

 

[pb4uski]

My planned AA includes 5.5% cash, and I am getting almost no yield from it.

Oh well. Equities go up and down, bonds yield more sometimes than others, and cash is yielding almost nothing. But you know, my AA was planned for the long term, and I haven't been chasing yield. I am no good at market timing so I just don't try to do it. My portfolio has been growing as a whole, and I am OK with my AA.

My target is 6% so we are similar in that respect. Usually I have about half in my Discover Bank FDIC insured savings account that earns 0.8% and the rest in the Vanguard Short Term Investment Grade Fund which has had a 0.85% return since I bought it in the spring of 2012 (good 3.78% return last year but taking it on the chin at -.86% so far this year).

 

[urn2bfree]

No. The reason for the cash bucket is 100% safety and liquidity, not to make a profit. Therefore it should be in something like short-term Treasuries.

How are Short term Treasuries at 0.2% yield any safer or more liquid than my credit union cash account at 1% yield? The safety seems equal and while not necessarily "profitable" I will take 1% over 0.2% or over a pointy stick in my eye anytime. I believe that slight but real difference could mean a difference in the impact of cash buffers on portfolio stability and survivability which these papers are questioning. At any rate the papers should use more realistic possibilities for safe, cash yields than whatever the LOWEST yield might be.

 

[teejayevans]

No. The reason for the cash bucket is 100% safety and liquidity, not to make a profit. Therefore it should be in something like short-term Treasuries.

I-Bonds work well (after the 1 year grace period), by definition they will keep up with inflation.
TJ

 

[rayvt]

How are Short term Treasuries at 0.2% yield any safer or more liquid than my credit union cash account at 1% yield?

Really? Osprey FlyerFDIC Seizes Missouri Bank, NCUA Closes 2 Credit Unions - Osprey Flyer
Granted, the FDIC/NCUA will eventually get you your money -- up to the insured limit.

In terms of safety and liquidity, US Treasuries are the gold standard. Everything else is lesser. Why do you think that the latest Treasury auctions were around $35 BILLION (not million, billion) per issue -- 4 week, 13 week, 26 week, 52 week, 5 year, 2 year? Treasury Auction Results

Now, you may prefer to reach for a bit more yield by going with a Credit Union, but don't kid yourself that it's the same quality.

Moreover, while individuals can get 0.8% to 1% in a bank/CU, these studies are for large groups of people, not individuals. No way can people on aggregate deposit hundreds of millions or billions of dollars in CUs.

I believe that slight but real difference could mean a difference in the impact of cash buffers on portfolio stability and survivability which these papers are questioning.

I doubt it would make any significant difference.

$1M portfolio, 4% SWR, 5 years in cash bucket = $240K in the cash bucket.
One-Year Treasury Constant Maturity yield is 0.11%. That's $264/yr on $240K. Pitiful.
Alliant CU pays 0.70%. That's $1680/yr
If you could get 1%, that's $2400/yr.

For an account the size of $240,000, the additional $1426 is so small it's invisible. Even $2200/yr is not going to make any difference in the portfolio stability and survivability.

 

[pb4uski]

.....Moreover, while individuals can get 0.8% to 1% in a bank/CU, these studies are for large groups of people, not individuals. No way can people on aggregate deposit hundreds of millions or billions of dollars in CUs....

Your argument above doesn't seem very relevant given the composition of this forum and the subject of the thread.

I would prefer FDIC insured accounts to Treasuries only because Treasury yields are so pathetic. IMO for individuals, FDIC insured savings accounts are better than short term US government full faith and credit instruments - it's the same counterparty backstopping the credit risk at the end of the day, the yields are better and there is no interest rate risk.

 

[karluk]

I doubt it would make any significant difference.

$1M portfolio, 4% SWR, 5 years in cash bucket = $240K in the cash bucket.
One-Year Treasury Constant Maturity yield is 0.11%. That's $264/yr on $240K. Pitiful.
Alliant CU pays 0.70%. That's $1680/yr
If you could get 1%, that's $2400/yr.

For an account the size of $240,000, the additional $1426 is so small it's invisible. Even $2200/yr is not going to make any difference in the portfolio stability and survivability.

Throwing away the chance to earn an extra couple of thousand dollars a year on the unsupported theory that it's "not going to make any difference in the portfolio stability and survivability" is not what I would call wise money management. It's similar to making the claim that investing in low cost Vanguard funds instead of other mutual funds with ~1% expense ratios won't have any long term effect on portfolio performance. A quick visit to Vanguard's web site should help educate you that small improvements in return, either through increased yield or lower expenses, will indeed dramatically improve long term returns and help insure portfolio stability and survivability.

 

[rayvt]

The Vanguard information is a non sequitur.

Sorry, but the data I have is directly on point and is why I said what I did. My claim is not unsupported.

I've run it using both 1-year Treasuries and 10-year Treasuries. The 10 yr, of course, yields a lot more than the 1 yr, and more than savings accounts or 6-month CDs. (You wouldn't use anything with a 10-year duration, though. Too much interest-rate risk.)

The 10-yr is yielding about 2% now. But it makes little difference in the results of a 30 year portfolio survival.

I'll be posting the link to my spreadsheet in a few days.

 

[calmloki]

I use cash to make money. We are still getting 4% on some PenFed Cd's, and make 0.75-0.9% on cash in accounts. The cash in accounts is used to fund property loans, which juice our overall cash accounts up to about 6+%. At any time we manage to have about 25% cash, 25% CD's, and 50% loans. The cash is very liquid, as opposed to the rental property, which really isn't.