Does a low-cost ETF outperform CEFs for current income?
In the discussion on a recent article reviewing a closed-end fund, income portfolio, one of Seeking Alpha's prolific commenters opined that simply putting money in Vanguard S&P 500 ETF (NYSEARCA:VOO) would "consistently" outperform. Unsurprisingly, the primary rationale came from the gospel according to St. Jack: Management fees will consume the ETF. Now, don't get me wrong here. I appreciate the value of simply holding a low-cost fund (or, preferably in my view, a portfolio of low-cost funds) as a wealth-building mechanism. But, as I suggested in my reply, the goals of a current-income portfolio - one that generates cash for living expenses - can be better met by closed-end funds.
For one thing the check is in the mail, on a monthly basis in many cases, without the investor having to do anything. To get comparable income from your low-cost ETF each month you have to sell shares. It incurs transaction costs, however trivial. It erodes share number, especially in times of declining markets making it harder to recover from inevitable downdraws. By contrast, closed-end funds supply attractive income with no effort, no transaction costs and no decrease in share number regardless of market conditions. That's not to say that there may not be capital erosion to NAV losses on occasion; of course there will be. But that's a fact of life with the ETFs as well. For CEFs the biggest threat to your income stream is a distribution reduction. That can happen, and its ever-present specter requires careful monitoring of the funds for looming red flags.
The comment, and my response, got me thinking on just how well the two approaches might compare. Rarely do I see analyses based on withdrawing income. More typically, writers discuss price return, dividend growth, or total return where distribution income is reinvested. So, I took a simple CEF portfolio that included two equity-income funds, one fixed-income fund and one preferred shares fund, and put it against VOO. The premise here is to start with the same capital, and withdraw income from VOO at an amount equivalent to the monthly payouts of the CEFs by selling shares.
I selected four widely held CEFs that are popular with many CEF investors. I did not try to cherry-pick funds, I selected funds that I have been writing about for years and have owned myself for several years. Here are the funds, which I'll consider in an equal-weighted, never-rebalanced portfolio.
- Eaton Vance Tax-Managed Div Equity Income (NYSE:ETY). I included this one because it was the commenter's example of a fund consistently underperforming VOO. It's one of Eaton Vance's option-income, equity funds.
- Eaton Vance Tax-Managed Buy-Write Opps (NYSE:ETV). Another EV option-income fund which I included because in my view it is the starter equity fund for someone entering the CEF space. Note that doesn't mean you should go out and open a position today, but that at attractive valuations, ETV should generate strong performance over an extended time frame with minimal drama.
- Flaherty & Crumrine Preferred Securities Income (NYSE:FFC). My favorite preferred shares fund. I've owned others but this is my core holding for preferred stock. I've written about it several times here at SA where I've laid out my reasons for this point of view.
- PIMCO Dynamic Income (NYSE:PDI). PIMCO is the master of generating high income from the debt and credit markets, and it's my strong opinion that PDI is PIMCO's best high-income fund. We can pick nits on that statement all day, of course, but PDI has the highest or close to the highest payout of all PIMCO CEFs. More importantly, it typically bounces between modest premiums and modest-to-significant discounts so it offers attractive entry points. This distinguishes it from many other PIMCO funds that consistently sell for unconscionable premiums.
To address (and concede) the commenter's primary point, here we have the cost of ownership for these funds (from cefconnect)
<Error Correction: Cefconnect is wrong on PDI's expenses. The correct value is a whooping 2.83%.>
I'll note that FFC and PDI are leveraged funds and these numbers include interest expenses (0.20 and 0.82%, respectively).
No matter how you slice it, VOO is the undisputed cost-efficiency champ. If your number-one, overarching, investment priority is to pay as little as possible to your investment manager, VOO is the way to go.
So, with that background, let's see what kind of numbers the two alternative investments have turned in. I'm going to restrict this to the past three years. That covers a terrific bull cycle for the S&P 500 which certainly plays in VOO's favor. But it also covers a smallish correction and a subsequent extended flat-market period, which should favor the fixed-income cohort. One of the things I like about the CEF approach is that it is somewhat better poised to handle market cycles without as much drawdown.
What I've done is to take the income generated by the CEFs and match it on a monthly basis by selling shares from VOO. VOO pays quarterly dividends, so in the months that VOO has a payout, shares are only sold to account for the difference between the CEF's and VOO's distributions. In the event VOO's quarterly dividend exceeded the CEF's combined distributions, the excess was to be reinvested, but that never happened. There's a telling observation in itself. VOO's quarterly dividend never exceeded a single month's distribution from the CEFs.
Comparing the Funds: Total Return
Starting with total return, here is how the individual funds stack up. Keep in mind that total return in this case assumes reinvesting the distributions as they are received, which we are not doing.
VOO has done well, but it's beaten only one of our four funds, ETY. The other three run about 10 points ahead for the three years. So much for "consistent underperformance." Let's put that misconception to rest.
Comparing the Funds: Income
But we are not reinvesting the income in this model, we're taking it out as cash as, say, a retiree will do. Here's where it gets interesting.
This chart shows annual income based on the distributions for the four CEFS from a portfolio that started at $100,000 on July 1, 2013 through June 30, 2016.
Note that the CEFs returned something close to 10% a year from distributions. By any measure, this is excellent income. To get a sense of where that income is coming from here are the current regular distribution yields for the four funds (from cefconnect).
How, you ask, can that generate 10% yields? Well, these regular distributions are not the whole picture. We're not seeing PDI's annual special distributions, which have driven its actual annual distribution yield well into double-digits. CEFs are required to return investment income to investors on an annual basis, so some funds need to make special distributions to meet this requirement. For 2015-16 PDI's special distribution was 0.99 times its average monthly distribution, doubling its yield. For 2014-15 it was 0.78 times the average monthly distribution, and for 2013-14, it was 0.48 times the monthly average. PDI's special distributions have been about the equivalent of adding the income from another CEF. Think of it as an end-of-year bonus.
To match that kind of income will, as you can imagine, take a massive chunk out of VOO each month. We're at more than double the widely recommended 4% withdrawal rate to maintain a sustainable current-income portfolio.
Comparing the Funds: Capital Sustainability
Over the past three years both the CEF portfolio and VOO have managed to sustain some capital growth while generating this high income.
For the bull market run early in the period, VOO grew at a greater pace than the CEF portfolio. However from the 2014 market peak, it has given up considerably more than the CEFs. At the beginning of July 2016, VOO grew to $102,184.47. For the same period, the CEFs grew to $105,755.31. I did deduct $7 per trade for commissions on the ETF sales, but for 36 months, this amounts to only $252 and it is a fixed cost, so for a portfolio starting at values of more than $100K it does not change. Furthermore, real Vanguard stalwarts are likely holding their funds in a Vanguard brokerage account anyway where the ETFs trade without transaction fees, eliminating this cost entirely.
Considering These Results
Frankly, I was surprised by these results. I had anticipated that VOO would lose value by keeping up with these high-income generators. Such is the power of that early bull market run. However, I'll note that VOO's share count declined by a split adjusted -21.2% from sales to generate that income. It is difficult to see that as a sustainable level of decay.
Three of the four CEFs managed to generate that income without losing value. FFC took a -5.9% decline. All four have exactly the same number of shares they started with. ETV, ETY and FFC paid exactly the same monthly distribution over the 36 months. PDI alone has increased its distribution from a starting point of $0.177/share to $0.221/share at present.
The decay in share number for VOO and the flat line for distribution yield for three CEFs does not auger well for prospects of income gain likely to keep pace with inflation. Such is typically the case for high-income portfolios. This is in contrast to a dividend-growth approach to income generation, but a DGI portfolio is more likely to be in the 4% income range than the 10% we see here. To offset the inevitable loss of income to inflation I set up my model CEF portfolio (here) to reinvest any distributions over 8% income returns and included funds that do have a history of raising distributions. That may not be sufficient to fully satisfy the inflation monster, but it should help generate some increases in income over time. It may make better sense to lower that number to 7% if inflation-proofing the income is a priority.
I'll be the first to admit that this is far from a definitive result, but as an example I believe it demonstrates how well chosen CEFs can be powerful generators of sustainable income. Moreover, they can do so more effectively than selling shares of a passive, indexed ETF. This should not be surprising; CEFs are investment tools specifically designed for income production. It's my view that a fund like VOO is an excellent choice for wealth building (as part of a diversified portfolio), but when one reaches a stage where the objective shifts to current income production, it only makes sense to investigate products designed with that objective at the forefront.
CEFs get little respect from many investors. I even had to sign a waiver with my broker acknowledging awareness that CEFs are high-risk investments before they would let me purchase these funds. At the time, it gave me pause and I had to decide if I really wanted to invest in CEFs. I can understand this lack of respect to a point. There are some poor choices in the CEF universe. Anyone who gets saddled with one of those poor choices will pay a price for that choice and may generalize the experience to all CEFs. CEFs take considerable research to do right. That research can be daunting to the CEF novice, but the reality is that there are few investment choices that do not require careful research. And, like any investment choice, there is a learning curve, so experience is rewarded. Despite the generally negative opinions many hold regarding CEFs, I submit that they are an excellent choice to provide steady, high income throughout a retirement. For the new investor, however, those caveats about research and knowledge borne from experience deserve careful consideration.
So, what about those fees and expenses? As I've been writing about CEFs here, the most frequent negative that readers put forward is their high fees. It is true that they are high, much higher than ETFs or even open-end mutual funds. But these funds are intensely managed with an sustainable high-income objective. That kind of management is necessarily expensive. As for any expensive service, one wants to be sure one gets value for that cost. I frequently omit any reference to fees when I write about CEFs. Not to avoid the issue, as some have suggested, but because I really don't pay much attention to them when I'm doing my research to the point that I hardly consider it as a metric. I look at performance data, yields, discounts, net investment income, distribution shortfall and the like. If metrics such as those check out, the fees are, in my view, earned, no matter what they are.
One last point on expenses: I'm not sure where it comes from (a misreading of Jack Bogle perhaps?) but I've found that many readers have the mistaken impression that fees come out of a distribution after it's declared and before the investor receives it. This recurs so often that I know it's common. It is, of course, completely wrong.
Disclosure: I am/we are long ETY, ETV, FFC, PDI.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I am not an investment professional and this article does not constitute investment advice. I am passing along the results of my research on the subject. Any investor who finds these results intriguing will certainly want to do all due diligence to determine if any security mentioned here is suitable for his or her portfolio.