7 Years' Living Expenses Largely in Short-Term Bond ETF: Thoughts?

[pb4uski]

On bond funds in general: What about a broadly diversified bond fund? Government and corporate aren't the only choices.

It depends on WHY you are owning those bond funds. If it is for diversification against stocks, just look at 2002 and 2008 to see that corporate bonds don't appreciate nearly as much, and sometimes drop, just when stocks are taken out and shot and/or there is a whiff of financial crisis in the air. Those are the periods where bond and cash diversification is usually most appreciated.

On the model: Long-term bonds is probably not the right asset class to be using in your tests. Use some type of intermediate bond index in your model. And if you still get the same results, it will be interesting.

It could well be that the cash position really helps when you have long bonds, but doesn't help so much when you have intermediate bonds.

I guess if Firecalc had such an option (intermediate term bonds) I would have used it but alas, it does not.

As I recall, government and corporates make up a large part of the bond market, at least Vanguard seems to think so. Unless you are differentiating government bonds from agencies, et al which is not a particularly useful distinction.

This fund is designed to provide broad exposure to U.S. investment grade bonds. Reflecting this goal, the fund invests about 30% in corporate bonds and 70% in U.S. government bonds of all maturities (short-, intermediate-, and long-term issues).

I'm not sure I agree with your last statement, but we'll agree to disagree.

The second test I did is probably more relevant to our discussion as the assets are a mix of a stock index and bond funds which is in line with your thinking, but still seems to have a minimal impact on the success rate, particularly if one makes some sort of mental adjustment for 1% interest on cash vs 0%.

 

[Cobra9777]

What is interesting is that I tried to test the notion that a cash buffer is suboptimal with Firecalc. Default assumptions except $1m portfolio, $40k withdrawals (4% WR).

Tried a run with a mixed portfolio of 5% US Small, 55% S&P 500 and 40% LT corporate bonds gets an 85.7% success rate. If I change the LT corporate bond to 34% and add 6% in cash (1 month Treasury) then the success rate INCREASES to 89.8%.

Surprised me and seems inconsistent with the paper's hypothesis.

I then did a run using the total market with 60% equities and the success rate was 95.6%. Then I reduced the $1m to $940k as if one had 6% in cash earning nothing and it reduced the success rate to 91.2%. This is probably too conservative as the cash would earn something (say 1% +/-).

This might be easier to model with ********, which has a cash category. I ran the same scenario as you... all default, $1M portfolio, $40K/yr spend, and 2 scenarios:

60/40/0... 90.43% success
60/34/6... 89.57% success

This seems more intuitive to me. It says you give up almost 1% success rate to "sleep better at night." I'm slowly coming around, and starting to question why I'm holding 5% cash.

 

[FUEGO]

I have 2 years in cash (which might get partially switched to CDs or bonds if rates ever dictate that's a sensible move).

That cash plus the dividends from my taxable account would get me through 2 years of ugly markets. Maybe it's psychological, but I feel more secure that way.

The other 97% of my portfolio is in equities, so any cash drag is minimal.

Looking back at the last three decades, I only see one occurrence of 3 years of consecutive annual losses (2000-2002). I'll have the flexibility to deplete my cash reserves during bad or flat markets (or keep the selling of shares to a minimum at least), except in those rare instances of down markets lasting 3+ years.

7 years of a very stable asset like short term bonds seems excessive and seems to represent 20% of the OP's assets that will produce inflation or inflation+1% each year. But hey, I'm not the one sleeping at night for him. Although as others indicate, he's possibly losing real security at the expense of false security.

 

[audreyh1]

The other 97% of my portfolio is in equities, so any cash drag is minimal.

Looking back at the last three decades, I only see one occurrence of 3 years of consecutive annual losses (2000-2002). I'll have the flexibility to deplete my cash reserves during bad or flat markets (or keep the selling of shares to a minimum at least), except in those rare instances of down markets lasting 3+ years.

7 years of a very stable asset like short term bonds seems excessive and seems to represent 20% of the OP's assets that will produce inflation or inflation+1% each year. But hey, I'm not the one sleeping at night for him. Although as others indicate, he's possibly losing real security at the expense of false security.

Yes, you can draw on cash or other fixed income to live off of. But you can also use it to buy more equities. People who rebalance their portfolios also use their cash and bonds to buy more equities when equities get hit hard.

And FWIW, we don't have to worry about the bond portion keeping up with inflation - it may not. That's why most of us hold equities in portfolios. And you don't have to hold a large portion in equities to keep up with inflation-adjusted withdrawals. As low as 40% equities does just fine over long periods.

You talk about "real" security, but models show that 100% stocks portfolios have poorer survival statistics over long periods than at least 20% fixed income. Just sayin'.....

Of course, if someone is close to 100% stocks because they are living off the stocks dividends and otherwise not selling investments in their portfolio, and they can survive a little dividend shrinkage during rough times, than the point is moot.

 

[pb4uski]

This might be easier to model with ********, which has a cash category. I ran the same scenario as you... all default, $1M portfolio, $40K/yr spend, and 2 scenarios:

60/40/0... 90.43% success
60/34/6... 89.57% success

This seems more intuitive to me. It says you give up almost 1% success rate to "sleep better at night." I'm slowly coming around, and starting to question why I'm holding 5% cash.

That seems more sensible to me and the sleeping better at night is well worth giving up 0.86% success rate.

Are you questioning holding 5% cash because the success rates are so similar? If so, I would flip it around.

 

[WannabeRetired]

That seems more sensible to me and the sleeping better at night is well worth giving up 0.86% success rate.

Are you questioning holding 5% cash because the success rates are so similar? If so, I would flip it around.

Yeah, that's the same way I viewed it, too. I'd feel more comfortable with a cash cushion than getting a tiny bit better success rate percentage.

 

[audreyh1]

I guess if Firecalc had such an option (intermediate term bonds) I would have used it but alas, it does not.

OK - I see what you are saying about Firecalc options, and it's too bad because many safe withdrawal rate models like Bengen's used "intermediate government bonds". And there is a huge difference in historical behavior between intermediate and long duration bonds.

 

[WannabeRetired]

<deleted>

 

[Mo Money]

Okay, given that things here in the off*ce are accelerating as I race to get tasks done in time for my departure at the end of this month, and based in part on the wisdom above, I sold a chunk of VCSH. Now I own: 1 years' living expenses in cash (earning 1%), plus 6 years of living expenses evenly divided into VCSH (2 years' worth of short term corp bond ETF); BSV (2 years' worth of short term bond ETF); and BIV (2 years' worth of intermediate term bond ETF).

In the end, as the cited articles (including Kitces) often point out, and as the posters above have often said or implied, my "bucket" of living expenses is a mirage/imaginary security blanket.

I'm fine with mirages and imaginary ways of seeing things -- -- if they help me feel secure, and are also rooted in reality when looked at dispassionately. My bucket/blanket of cash and bonds is really just a ~23% slice of my portfolio. The balance of my holdings are in equities. So I am essentially a couch potato 75/25 investor. As long as I aim to stick at that allocation, then bucket or not, I am going to be rebalancing periodically as my 75/25 gets out of whack. If times get hard, I can cut my living expenses by probably one-third, and I can always work part-time. But with 35 years of expenses in a 75/25 portfolio, I should be okay.

Thanks for the insights!

 

[audreyh1]

Yeah - with only 25% in fixed income, you don't have to worry about "drag" at all IMO. You just have to be able to stomach the occasional roller coaster! LOL!

Congrats on reading all that material and absorbing it so quickly!

 

[Chuckanut]

One thing that must be considered is the human emotional element.

If one's ability to act rationally in the face of a relentless bear market is strong, then keeping a big cash buffer is not necessarily useful.

If one is going to panic and sell low, then having the extra cash security blanket that will reduce or eliminate panicky selling, is a good idea.

Know thyself. Just my 2 cents.

 

[Cobra9777]

That seems more sensible to me and the sleeping better at night is well worth giving up 0.86% success rate.

Are you questioning holding 5% cash because the success rates are so similar? If so, I would flip it around.

I've been questioning my 5% cash allocation for some time, mainly due to the drag on performance. My total return in 2014 was 9.3%. But excluding cash, it was 9.8%.

Also, we have a very low withdrawal rate. We have pensions and rentals that cover 70% of expenses. Cash dividends from the taxable account cover another 15%, with SS still to come. In a down year, we could cut travel and we're covered.

Many of my bond ETFs have significant cash positions, so my actual cash allocation is much higher than 5%.

Finally, I'm warming up to the concept that selling equities in a downturn is not the end of the world if you immediately rebalance and do it tax efficiently. You have to replenish the cash anyway and that means selling something. Why not just skip the cash and sell?

Yes, given the myriad uncertainties inherent in any 30-40 year retirement plan, a 1% reduction in success rate seems quite trivial. But just add it to the list. And the more I think about the list in its entirety, the less compensated I feel for the lower success rate and lower performance.

I'm still holding cash for now as I'm only 1.5 years into ER and still learning the ropes. But I'm definitely warming up to the arguments against holding cash.

 

[audreyh1]

I've been questioning my 5% cash allocation for some time, mainly due to the drag on performance. My total return in 2014 was 9.3%. But excluding cash, it was 9.8%.

Also, we have a very low withdrawal rate. We have pensions and rentals that cover 70% of expenses. Cash dividends from the taxable account cover another 15%, with SS still to come. In a down year, we could cut travel and we're covered.

Many of my bond ETFs have significant cash positions, so my actual cash allocation is much higher than 5%.

Finally, I'm warming up to the concept that selling equities in a downturn is not the end of the world if you immediately rebalance and do it tax efficiently. You have to replenish the cash anyway and that means selling something. Why not just skip the cash and sell?

Yes, given the myriad uncertainties inherent in any 30-40 year retirement plan, a 1% reduction in success rate seems quite trivial. But just add it to the list. And the more I think about the list in its entirety, the less compensated I feel for the lower success rate and lower performance.

I'm still holding cash for now as I'm only 1.5 years into ER and still learning the ropes. But I'm definitely warming up to the arguments against holding cash.

Sounds like your goal is maximizing the size of the portfolio say 10 years down the road? Or maximizing what you leave to heirs?

For some folks, having a larger portfolio is their goal, so they are willing to make sacrifices - lower withdrawals today, belt tightening during market downturns, etc., to accomplish that goal.

Some of us are more interested in spending (and gifting) more in the early years and don't plan to leave a lot after we pass. We just want the portfolio to survive during our lifetime, and prefer to live with less year-to-year volatility.

It really depends on your goals.

 

[Texas Proud]

OK, I want to know how people who are doing this X years of expenses set aside actually handle this....

IOW, say you have 3 years of expenses set aside in a MM account or a CD ladder.... now we go one year into the future... do you now only have 2 years If so, does it not worry you that you have spent 1/3rd of your bucket If you have replenished it, then what benefit have you gained

It just does not make sense to me to hold so much out of the market... we just lived through probably the 2nd worse market decline in American history and the market was back in 5 years... so, you might say 'yes, but I did not have to sell when it was down'... but when did you make the decision to fill your bucket back to full

This seems like a backwards way of timing the market.... and that a rebalance would be better....

 

[FUEGO]

OK, I want to know how people who are doing this X years of expenses set aside actually handle this....

IOW, say you have 3 years of expenses set aside in a MM account or a CD ladder.... now we go one year into the future... do you now only have 2 years If so, does it not worry you that you have spent 1/3rd of your bucket If you have replenished it, then what benefit have you gained

I'm holding cash in a money market at my credit union and in a single bond that matures soon.

I'm getting down to the level where I want to refill the cash bucket so I'm about to put in an order to sell $10000 or so of a particular ETF I want to get rid of.

In six months, I'll sell another $10,000 or so (or whatever I need to top off the cash reserve account) if the market is reasonably steady (flat to up). Otherwise hold off. I'll admit to muddling through this and that there are some market timing elements to this little experiment. I can't imagine market timing on under 1% of our portfolio having a statistically significant impact on our portfolio survival numbers.

 

[jabbahop]

This might be easier to model with ********, which has a cash category. I ran the same scenario as you... all default, $1M portfolio, $40K/yr spend, and 2 scenarios:

60/40/0... 90.43% success
60/34/6... 89.57% success
----

thanks for running this. Isn't 6% 1.5 years of cash?

I've been mulling how much I want in cash as well. Using ******** ...

Using the $1M portfolio and $40K withdrawals; rebalancing on.

a) 5 years of cash - 60% equities; 20% bonds; 20% cash (40K x 5 years)

b) 1 year of cash - 60% equities; 36% bonds; 4% cash (40K/1M)

a) 80% success rate; $633K median ending portfolio value
b) 90% success rate; $841K median ending portfolio value

did I get something wrong here?

 

[FUEGO]

You talk about "real" security, but models show that 100% stocks portfolios have poorer survival statistics over long periods than at least 20% fixed income. Just sayin'.....

Of course, if someone is close to 100% stocks because they are living off the stocks dividends and otherwise not selling investments in their portfolio, and they can survive a little dividend shrinkage during rough times, than the point is moot.

Eh, not a big deal really. I took a look at a 3.5% withdrawal rate ($35,000/yr on $1 million). 75% stocks/25% bonds = 99% survival rate, 100% stocks/0% bonds = 98% survival rate.

I'm losing 1% survivability in exchange for having roughly twice as much at the end of 50 years (I'll only be 84 at that point). $5 million bucks for a 1% difference in portfolio survivability? A calculated risk I'm willing to take.

Part of my rationale is that we won't follow the standard "spend 3.5% of initial plus annual CPI increases". It'll be some variation of a variable spending rate based on portfolio value each year. I'd rather have a standard of living that increases slowly over time (due to appreciation of equities). That comes at the price of risking some years of limiting spending to a bare bones budget if we see a Great Depression (a plight somewhat mitigated by our cash reserves).

 

[2B]

Of course, if someone is close to 100% stocks because they are living off the stocks dividends and otherwise not selling investments in their portfolio, and they can survive a little dividend shrinkage during rough times, than the point is moot.

The problem with living off the dividends is that financial stock dividends were all but eliminated in 2008 - 2010 and they haven't fully recovered. A "dividend portfolio" in 2007 would have been heavy in financial stocks including the now all but worthless Fannie and Freddy common and preferreds.

 

[audreyh1]

OK, I want to know how people who are doing this X years of expenses set aside actually handle this....

IOW, say you have 3 years of expenses set aside in a MM account or a CD ladder.... now we go one year into the future... do you now only have 2 years If so, does it not worry you that you have spent 1/3rd of your bucket If you have replenished it, then what benefit have you gained

It just does not make sense to me to hold so much out of the market... we just lived through probably the 2nd worse market decline in American history and the market was back in 5 years... so, you might say 'yes, but I did not have to sell when it was down'... but when did you make the decision to fill your bucket back to full

This seems like a backwards way of timing the market.... and that a rebalance would be better....

You withdraw from the portfolio annually. Part of that withdrawal replenishes your short-term cushion. Say after one year, what was three years expenses went down to two, then your withdrawal gets it back to three.

Now if it was a particularly bad market year and the portfolio dropped, you have some options. You can withdraw less from the portfolio and use the cash cushion to make up part of the difference if you like. You have flexibility in how you approach it. Those of us using the % of remaining portfolio withdrawal method will occasionally have to deal with a substantially lower $ withdrawal from the portfolio after a bad market year.

People for whom "It just does not make sense to me to hold so much out of the market" are looking to increase their long term return. Some of us don't care about that, and aren't looking to increase their portfolio volatility. We'd rather have some cash available for spending in the short term. These are simply different goals.

 

[audreyh1]

The problem with living off the dividends is that financial stock dividends were all but eliminated in 2008 - 2010 and they haven't fully recovered. A "dividend portfolio" in 2007 would have been heavy in financial stocks including the now all but worthless Fannie and Freddy common and preferreds.

Perhaps why I have never been comfortable taking that very tax efficient approach, but a lot of people do take that approach and feel comfortable with it. And I thought stock dividend ETFs like VIG and VYM did just fine recovering from that period. Maybe not DVY.

 

[Cobra9777]

thanks for running this. Isn't 6% 1.5 years of cash?

I've been mulling how much I want in cash as well. Using ******** ...

Using the $1M portfolio and $40K withdrawals; rebalancing on.

a) 5 years of cash - 60% equities; 20% bonds; 20% cash (40K x 5 years)

b) 1 year of cash - 60% equities; 36% bonds; 4% cash (40K/1M)

a) 80% success rate; $633K median ending portfolio value
b) 90% success rate; $841K median ending portfolio value

did I get something wrong here?

Your 5-year scenario used the default 0.25% return on cash. That's not realistic as most people holding 5 years cash would have some kind of CD ladder or ST bond fund. Substituting 1.5% in that scenario increases the success rate to 87%... still not radically different from the 1-year scenario at 90%.

Also, you have to consider other sources of cash such as pensions, SS, etc. If nothing else, $600K in equities will spin off ~$12K dividends. So your "years of cash" and the resulting AA seem a little high to me. In my case, 5% cash covers about 3 years.

 

[pb4uski]

OK, I want to know how people who are doing this X years of expenses set aside actually handle this....

IOW, say you have 3 years of expenses set aside in a MM account or a CD ladder.... now we go one year into the future... do you now only have 2 years If so, does it not worry you that you have spent 1/3rd of your bucket If you have replenished it, then what benefit have you gained

It just does not make sense to me to hold so much out of the market... we just lived through probably the 2nd worse market decline in American history and the market was back in 5 years... so, you might say 'yes, but I did not have to sell when it was down'... but when did you make the decision to fill your bucket back to full

This seems like a backwards way of timing the market.... and that a rebalance would be better....

In my case at least and many others I suspect, I'm not holding money out of the market because my stock portfolio is still 60% of the total. What I am doing is holding a portion of my fixed income portfolio in cash.

I concede it is suboptimal. IIRC Vanguard says that the 10 year treasury is a reasonable proxy for the next 10 years bond returns, so let's say that is 3% and I earn 0.9% on my cash. The overall affect on the portfolio is roughly (3% - 0.9%) * (6% + 4%)/2 or 0.1%. The second part of the equation reflects that over the course of a year my cash declines from 6% at the beginning of the year to say, 4% at the end of the year because of withdrawals during the year.

I also concede it is a bit silly but having that cash at my beck and call makes me comfortable, sort of like an emergency fund and costs me very little.

I replenish the 6% annually when I rebalance.

 

[Live And Learn]

I'm still holding cash for now as I'm only 1.5 years into ER and still learning the ropes. But I'm definitely warming up to the arguments against holding cash.

I currently have 6 years in cash and short term bonds (3 years cash, 3 years ST bonds). I'm always questioning myself on this. When I calculate my WR, or use FIRECalc or some other calculator I totally ignore the cash portion (in other words, I pretend the money just doesn't exist).

While I claim to not be a market timer I am looking for a clear buy signal to put 3 years in either intermediate bonds or equities.

Clearly I am hopeless.

 

[audreyh1]

While I claim to not be a market timer I am looking for a clear buy signal to put 3 years in either intermediate bonds or equities.

Clearly I am hopeless.

Honestly, I don't think those clear buy signals exist. Of if they do occur, the market has already run up in anticipation.

 

[Options]

People for whom "It just does not make sense to me to hold so much out of the market" are looking to increase their long term return. Some of us don't care about that, and aren't looking to increase their portfolio volatility. We'd rather have some cash available for spending in the short term. These are simply different goals. [Emphasis Added]

+1
It's easy to ignore the psychological/behavioral aspect of PF management during a bull market, particularly one like we've been in for the past several years. It was Bernstein (among others) who said people discovered how much they had vastly underestimated their risk tolerance during the 2008 downturn. If others, with decades more investing experience either capitulated or came *this* close to capitulating at that time, what makes me think I wouldn't? AFAICT, in investing we are our greatest enemies and nothing is more dangerous to PF survival than ignoring behavioral risks.

I like SWAN portfolios. Count me in the reduced volatility/short term cash camp.