spending portfolio - selling versus dividends

[JohnEyles]

Ok, another basic question - but maybe it'll generate some interesting discussion.

I'm starting to think building a retirement nest egg is the easy part of "investing"
(not counting the having to go to work part) - you just put everything into some
nice mix of stock index funds, save a year or so of expenses in cash equivalenets,
and check your balances on the computer when you oughta be working !

Spending it down, in retirement, is the complicated part. What I can't quite get
my head around (and I always see it mentioned tangentially, and not directly
addressed) is to what extent do you generate your 4% (or whatever) WR by
selling off shares of investments, and to what extent do you generate it by
moving into higher-yielding investments and harvesting the dividends and
interest ?

In terms of portfolio survivability versus a certain WR, all that really matters is
what the *total return* of your portfolio is versus your WR and the rate of
inflation (and of course the timing of the total return, when the up and down
years occur). So to that extent it doesn't matter whether the total return is
from appreciated share values or from dividends. Is one somehow superior to
the other, nonwithstanding 2nd-order effects like low cap-gains rates in taxable
accounts ?

In the bucket framework, I suppose the answer is that the question is moot for
bucket 1 - it's just M'Mkt, CDs, short-term Treasuries and such. And bucket 3
should simply be for the best long-term total ROR. But what about bucket 2 ?
Looks like most people want a mix of longish-term Treasuries, other bonds, and
balanced income funds. I guess the idea would be to sweep the dividends and
interest into bucket 1, but also sell shares when 1 drops too low (presumably
the income from 2 would not be enough to keep 1 full).

As some have said, the balanced funds seem problematic, because you can't
tease out the stock and bond components to be sold separately. But their mix
of high income and stable but appreciating share price is appealing. I wonder if
it's possible to duplicate VWINX with one stock fund plus one bond fund ?

 

[brewer12345]

Hmm, well, I would say that return is return, regardless of whether it is from cap gains or portfolio income. Having said that, you are likely to find that the return of stuff that spits out significant cash is less volatile over time.

What I think most people do is spend dividend and interest income from the portfolio, then sell off small chunks of stuff that has risen, whether explicitly or via asset allocation. Obviously, if your portfolio income meets all your needs, then all you need to do is trim and rebalance from time to time. But if portfolio cash flow doesn't entirely cut it, then you can simply raise cash as you do your annual rebalancing.

WRT balanced funds, the whole point is that you don't need to tease out the stock and bond components separately. Such funds do the rebalancing for you. But yes, you could essentially replicate a baanced fund by holdng the appropriate % of a bond fund and an equity index fund. FWIW, I think such a portfolio would be foolishly undiversified.

 

[d]

I wonder if it's possible to duplicate VWINX with one stock fund plus one bond fund ?

i suspect one could come close ... i'd start by looking at value index plus total bond, but you would have to replace the management skill that wellington management brings to the table ... doub't that i can do as well as can they.

 

[brewer12345]

you would have to replace the management skill that wellington management brings to the table ... doub't that i can do as well as can they.

I have a great deal of respect for Wellington (and Dodge & Cox), but I have a hard time believing that the asset bloat they have undergone allows them anything like the freedom of action they had before. I think the bloat dilutes the skill pretty heavily.

 

[JohnEyles]

... you are likely to find that the return of stuff that spits out significant cash is less volatile over time.

Arguing for growth stock funds in Bkt3 and income funds (stock and balanced, in
addition to bonds/bond funds) in Bkt2 ?

What I think most people do is spend dividend and interest income from the portfolio, then sell off small chunks of stuff that has risen, whether explicitly or via asset allocation. Obviously, if your portfolio income meets all your needs, then all you need to do is trim and rebalance from time to time. But if portfolio cash flow doesn't entirely cut it, then you can simply raise cash as you do your annual rebalancing.

Yes, but this kinda begs the question of to what extent do you put money into
income-producing investments, and to what extent do you go for share appreciation
(is this the same thing as "growth" ?) and plan to sell shares ?

... you could essentially replicate a baanced fund by holdng the appropriate % of a bond fund and an equity index fund. FWIW, I think such a portfolio would be foolishly undiversified.

Agreed. My notion is that a balanced fund is mainly in Bkt2, and just a portion of
that, in addition to bonds and bond funds - something like AGG, and individual TIPS
(as you suggested in another thread).

 

[d]

the bloat dilutes the skill pretty heavily

nonetheless, i doubt i could do as well

 

[brewer12345]

Arguing for growth stock funds in Bkt3 and income funds (stock and balanced, in
addition to bonds/bond funds) in Bkt2 ?

Yes, but this kinda begs the question of to what extent do you put money into
income-producing investments, and to what extent do you go for share appreciation
(is this the same thing as "growth" ?) and plan to sell shares ?

Agreed. My notion is that a balanced fund is mainly in Bkt2, and just a portion of
that, in addition to bonds and bond funds - something like AGG, and individual TIPS
(as you suggested in another thread).

I "don't do" buckets (I think it is a fad/scam/bad idea), so I can't help you there.

Like I said, I don't think the source of the return matters (income or cap gains). So I don't think it is smart to throw your asset allocation out the window just because something has a high yield (otherwise you'd be 100% long stuff like EGLE, Canroys, mortgage REITs and other stuff with double digit yields). I think the smarter way to do this is arrive at an asset allocation that gives you a reasonable combination of expected return, volatility, and diversification. Then you use whatever cashflow the portfolio throws off for spending, supplemented as necessary with some cap gains skimmed off the top when you rebalance.

 

[Nords]

Spending it down, in retirement, is the complicated part. What I can't quite get
my head around (and I always see it mentioned tangentially, and not directly
addressed) is to what extent do you generate your 4% (or whatever) WR by
selling off shares of investments, and to what extent do you generate it by
moving into higher-yielding investments and harvesting the dividends and
interest ?

A theme on this board is that nearly every Young Dreamer thinks ER is a huge challenge... until they ER. And every Young Accumulator thinks the distribution phase is a huge challenge... until they ER.

For everyone else there's paralysis by analysis!

One of the more popular columns a year or two ago opined that a portfolio consumed during retirement could be built at 25x expenses (hence the 4% SWR). Building a portfolio that generated all its income through dividends, though, would require approx 33x expenses (a dividend yield of 3.3%). If you can find a higher-yielding investment, of course, then you'll need less of a portfolio. Good luck with that.

I think the whole spending concept is being overcomplicated and the buckets aren't helping. Pick an asset allocation. Rebalance it once in a while. When you're retired, rebalance in a way that maintains the asset allocation while coughing up some spending cash. The system works pretty well with or without buckets and whether you rebalance annually, every 57.3 months, or every other Thursday.

In our case we keep two years' expenses in cash and the remainder in equities. Our Tweedy, Browne Global Value fund soared over the last few years and exceeded its asset allocation by at least 5% of the portfolio's total size. So last year we sold off a year's worth of spending cash and later sold some more of the fund to buy a much cheaper index fund (PowerShares International Dividend, PID). We'll just keep reinvesting those dividends until we get our asset allocations where they need to be. We also reinvest the dividends from our small-cap value ETF (IJS) and the DOW dividend ETF (DVY). Dividend-paying funds are about 40% of our portfolio.

This year Berkshire Hathaway has run up to just over 30% of our portfolio (I hope it's not done yet!) and should probably be whittled back to 25%. We'll think about selling some of it next year but we could also do much better on expenses by liquidating what's left of the Tweedy fund (which is up nearly 12% YTD). We're not chasing dividends or growth right now and we'll probably stick to something near our current asset allocation. If I had to predict which investment is doing better next year, Buffett or a bloated expensive mutual fund, then I'd sell more Tweedy.

 

[kcowan]

Yes you have to get away from the idea of owning shares. Think about renting them for a while. Collect the rental returns by rebalancing. If you are in a cash squeeze (unexpected expenses), sell your winners or your dogs depending on your rebalancing rules. Usually allocation rules will have you sell your winners so just keep the cream and defer the rebalance reinvestment for a while.

These rules work over the long term by imposing discipline in place of emotion but it is OK to have short term imbalances as cash needs arise. Otherwise you will be forced to sit on more cash than you should.

 

[audreyh1]

Here's how some folks do it in a taxable portfolio:

Mutual funds throw out distributions which are taxable. These are both capital gains distributions and dividend type distributions. It's a good idea to take these as cash (i.e. not automatically reinvest) since you have to pay taxes on these distributions. If these come out to about 4%, you've got your withdrawal. If not, then sell some of the best performing asset classes (which will incur more taxes). If more than 4% then invest the "extra" in the worst performing asset classes.

This is simply how you withdraw and rebalance with minimal tax impact.

A tax-deferred portfolio would use a different approach since you only incur taxes on money you withdraw from the portfolio. You can rebalance the portfolio at any time without worrying about the tax consequences.

Audrey

 

[JohnEyles]

I "don't do" buckets (I think it is a fad/scam/bad idea), so I can't help you there.

I see your point; but maybe think of buckets as simply a useful tool for deciding how
much of your portfolio to allocate to cash equivs and bonds and conservative equities.
Seems like a better way of thinking of it than just some general notion that 60/40
should be right. The $10-millionaire without extravagant living expenses would
have way too conservative a portfolio at 60/40, don't you think ? - whereas
buckets would give more sensible allocations.

... arrive at an asset allocation that gives you a reasonable combination of expected return, volatility, and diversification. Then you use whatever cashflow the portfolio throws off for spending, supplemented as necessary with some cap gains skimmed off the top when you rebalance.

Wow, that's the clearest summary I've ever seen of how to think about drawing down
a portfolio. Nice.

Maybe I'll think about stopping all my DRIPs. (Makes computing capital-gains a LOT
easier too !)

 

[bosco]

I got ahold of "Buckets of Money." I have read about 80% of it.

I am not impressed. It's a marketing concept, not an investment concept as far as I can see. I have found almost no useful information in the book. Yes, 3 buckets are defined, but there is so much vagueness and flexibility in how money is or is not moved between them that, imo, the book degenerates into a D+ investment book. From a practical point of view, buckets = asset allocation. By the time you "borrow some of bucket 2 to get good buys in bucket 3, but be sure to pay it back" etc., you are asset allocating with optional market-timing thrown in.

Read 4 Pillars.

Some guy invented some catch-phrases, repackaged conventional knowledge, and has convinced a bunch of people it is something new. But plow past the fluff and fizzle and there's not much there.

Anybody want a "Buckets of Money" book for $16?

 

[youbet]

By the time you "borrow some of bucket 2 to get good buys in bucket 3, but be sure to pay it back" etc., you are asset allocating with optional market-timing thrown in.

Read 4 Pillars.

Yes, what you said!

But there was a plus in reading Ray's book for me. It caused me to rethink my plans and outlook in managing my portfolio in RE. By the time I analyzed what Ray was actually doing in terms of allocation, rebalancing, timing and withdrawal strategies, I had a little deeper insight into the strategies I developed over years and my own plans. I wound up reaching to the bookshelf for "4 Pillars" and several of the other classics to confirm what I was thinking regarding what Ray's new fangled jargon was actually saying. And that led me to a few minor tweaks.

Analyzing so-called new ideas to see if they are really new can be a mind stimulating process.

The biggest negative I have with Ray is that he's added new jargon to already existing strategies so that to understand what some Ray-fans are doing, you have to go learn the language. I might say that because equities have significantly run up, I've sold some high fliers and rebalanced back towards my target allocation percentages. Bucketizers might say they took advantage of high equity prices to sell some of bucket three to refill bucket two. We both did the same thing for exactly the same reasons, but you've got to interpret the jargon to see that. Kind of a waste of time.