Many months ago, I promised to look at the practicalities of using income-seeking ETFs to generate an income in retirement.
It seemed sensible to split the post into two parts:
- The first, already published, looked at purely passive income-seeking ETFs.
- A second post – this present one – looking at so-called Smart Beta income-seeking ETFs.
That said, as I’ve remarked before I’m not necessarily a fan of ‘smart’ income-seeking ETFs. Countless investors had their fingers burned investing in the popular iShares’ FTSE UK Dividend Plus ETF (Ticker: IUKD) ten years ago, having mistakenly seen it as a lower-cost and less-risky way of getting an income from higher-yielding shares.
I’ve also revised the intention of this post. As I was writing it, I increasingly saw less point in trying to construct an example portfolio, as I did with those purely passive ETFs.
There are so many choices to make – many of them very personal – and the chosen Smart Beta methodology plays a critical role. My own selection would arguably be arbitrary compared to yours.
Rear view mirror
Smart ETFs aren’t smart in the sense that a real-life manager of a fund or investment trust might be. There’s no judgement at work, or years of investment experience.
Instead, a theoretical index is constructed containing shares that meet specified criteria – criteria that for the purposes of this post would supposedly correlate with high, robust, and reliable levels of income.
By definition, then, these smart income approaches necessarily look backwards. They extrapolate into the future a set of dividend-paying characteristics that have been observed in the past.
And the past, as the financial services industry is duty bound to remind us, is not a reliable indication of what may happen in the future.
The consistent growth approach
A number of ETF providers offer ETFs that aim to track S&P’s family of Dividend Aristocrat indices. These contain shares with a history of increasing their dividend payments for 25 consecutive years or more, subject to a (large-ish) minimum market capitalisation. The final decision on inclusion is made by a committee.
On the face of things, this is a reasonable approach. Not least because a consistently growing dividend may speak to substantial free cash flow, which is always a good thing.
But as an income investor, I know all too well that there is no guarantee that a company that has grown its dividend for 25 years or more will continue to do so. Or indeed, that it will continue to pay dividends at all.
Consider Tesco, for instance. The grocery giant abruptly cut its dividend by 75% in August 2014, and abandoned paying it altogether in January 2015 — and it, too, had that golden 25 year reputation.
More recently, America’s gigantic GE—a member of the Dividend Aristocrat index for 35 years—was booted out after incoming chief executive John Flannery cut the dividend. Pfizer offended similarly.
Perhaps more fundamentally, the Dividend Aristocrat approach excludes any consideration of yield. As an income investor, I want consistency, but I also want income.
Granted, the Dividend Aristocrat approach has been shown to deliver superior overall returns, but at the end of the day, we’re talking income here.
The higher-yielding approach
An alternative approach goes gung-ho for income, building indices of higher-yielding shares.
Vanguard’s High Dividend Yield ETF, for instance, tracks the FTSE Russell’s matching FTSE All-World High Dividend Yield Index. MSCI also has its own High Dividend Yield Index, while Dow Jones maintains a series of Select Dividend indices.
iShares ETFs often (but not exclusively) use the Select Dividend series, while providers such as WisdomTree and Amundi often use the MCSI indices.
STOXX and Société Générale maintain their own flavours, the latter under the Quality Income index. These are less widely available than the majors – and when ETF market capitalisation has a bearing on charges, that can be a disadvantage.
- To see the wide range of dividend-touting ETFs available, check out this list from the JustETF tool. We only cover a taster of what’s out there in this article. Be sure to research exactly what you’re buying if you choose to invest!
Methodologies vary. FTSE Russell’s approach is to capture the highest-yielding 50% of the market, ranked by forecast yield, for instance. Others are pickier: STOXX wants to see positive dividend growth over five years, and a dividend payment in four of the last five years, while Société Générale adds nine ‘quality factors’ to the mix, including profitability and solvency.
So how to pick an ETF constructed on ‘higher yielding’ basis? My suggestion would be to start with the index methodology, and choose the one with which you are most comfortable. Which comes closest to your own stock-picking approach, or seems the most sensible?
Don’t be surprised if that leads to some awkward decisions. The iShares STOXX Global Select Dividend 100 ETF (Ticker: ISPA), for instance, seems only to be available on the German bourse with some brokers, while hunting for Lyxor’s SG Global Quality Income ETF at your broker may take you to Paris (although it is available on the LSE under the ticker SGQP). From a ‘smart’ perspective, both of these seem to me to be ‘smarter’ than ETFs based (say) on the more simplistic FTSE Russell approach.
But if you look to buy on costs – as many people do with ETFs – then Vanguard’s FTSE Russell-based ETFs are undeniably considerably cheaper than Lyxor’s SG Global Quality Income and iShares’ STOXX offerings.
What would I do?
On the one hand, you have fairly sophisticated products based around Société Générale’s Quality Income index methodology, and on the other you have the broad simplification of the FTSE Russell approach. Both have their merits.
Which is best? There isn’t a simple answer.
On the whole, I remain to be convinced of the worth of smart income-focused ETFs – compared to say income investment trusts.
None of these Smart Beta products feature in my own portfolios, nor are likely to.
But if I were to seek to build a retirement income from such a set, I’d want to spread my risks by opting for multiple providers, and multiple methodologies. Nor would I buy solely on cost – diversification has a price, and it’s probably one worth paying.
Why the emphasis on diversification? Simple: to try to minimise the downsides of ETF algorithms blowing up, à la IUKD.
Finally, I know that I’ve been away from the site for many, many months. Sorry about that: sometimes, real life gets in the way.
But I know – from the nudges I’ve had from The Investor – that at least a few of you have missed me. It’s good to be back.
You can catch up on all Greybeard’s previous posts about deaccumulation and retirement.
Comments on this entry are closed.
I am indeed very glad to have The Greybeard back. 🙂
Polite reminder: we’ve had huge debates arise multiple times on these ‘income in retirement’ posts from that faction who believes it’s best to focus on total return and to sell units, rather than to pursue an explicit income policy in retirement.
We get it. My own co-blogger thinks that.
However I would like any comments on this post to focus on at the very least retirement income from investments, and preferably from passive investments or at the least comparisons with investment trusts. Debating whether or not you’d want to do that is for another day/article.
Here’s a post I wrote when things got out of hand last time: http://monevator.com/weekend-reading-a-new-comment-policy-to-protect-on-topic-discussion/
As discussed in that post, anything I consider off-topic is liable to be deleted.
It’s much more useful for readers who are interested in the subject at hand to be able to hear some more angles on it than for people who are against the whole idea to repeat the fact again.
Thanks! 🙂
Great to see The Greybeard back, I’ve been looking forward to some more of his wisdom.
Very interesting as I, like some of the Monevator audience am in the process of transitioning to “retirement” whatever that may be.
I have been experimenting with VHYL and also Vanguard’s FTSE Equity Income funds as a diversified way of generating income.
I’d love more on Investment Trusts as I still find them quite opaque and hard to research. Especially against the above funds. Particularly interested in the hard evidence behind the common belief that they are a better port in a market storm especially from an income perspective considering their higher costs. We haven’t had such a storm for a good while and most of the research data re the funds vs trusts doesn’t seem to go back that far or the funds have not existed long enough.
Also interested in what possible income in retirement portfolios would look like vs the many different ones on here for the accumulation phase. I’m just starting to work my way through Michael McClung’s Living Off Your Money.
Finally can I put in a request for a little more on bonds which I am still struggling with – e.g. If US treasuries are now yielding 3% and above would you not buy a hedged US bond fund (does one exist?) over UK gilts? Possibly a little off topic for this conversation but extremely valid from an retirement income perspective as a good proportion of any retirement portfolio is likely to be made up of bonds.
Thanks!
Welcome back Greybeard and thanks for such a useful post.
Perhaps worth adding in terms of cost that many ETFs do not attract stamp duty and therefore have a 0.5% advantage over ITs from the start. Perhaps not so significant if investing small sums, but for a large portfolio could be a significant saving.
Have been looking carefully at some of the WisdomTree offerings – their smallcap dividend series has performed well and might be of interest for those interested in both income and growth.
Agree that a post on Bond ETFs or funds would be welcome since these will often form part of a passive portfolio.
Happy to see Gb back.
“would you not buy a hedged US bond fund (does one exist?) over UK gilts?”
I’d certainly be interested in TIPS rather than Index-Linked Gilts. Do any other countries offer decent yields on their inflation-protected sovereign debt?
I have a fair slug in these ETFs, split mainly between SPDR Dividend Aristocrats and iShares (and some in VHYL) and covering various regions to get global coverage. For the reasons Greybeard states I would never rely on any one methodology or provider, let alone a single ETF.
My natural yield income strategy needs closer to a 4% yield rather than the measily 2% yield of a global index ETF like VWRL.
I also have several investment trusts targeting similar yields as an overall average. Again, it means no single point of failure or underperformance, and payouts spread across the year. I don’t care if there is sometimes an overlap in stocks held.
The UK Dividend Aristocrats ETF hasn’t done great in the last 5 years. FTSE100 index ETFs have killed it in total return and still yield more (with lower charges). In contrast IUKD has a healthy yield at present.
My experience suggests that the approach can be profitable but whether such an approach is a good idea now, I have serious doubts.
Equity income as an approach has frequently been unloved but when market yields fall, as they have in recent years the market for income producing products grew substantially . Companies have moved their dividend policies accordingly, giving the market what it asks for and performance of recent years had been disappointing.
I wonder that when an approach becomes popular, we’ll known and probably distorted then underperformance is not surprising…
“For example, there are three versions of the S&P 500 index: price return, which only considers the price of the components, total return, which accounts for dividend reinvestment, and net total return, which accounts for dividend reinvestment after the deduction of a withholding tax.”
https://en.wikipedia.org/wiki/Stock_market_index
Life would in some ways be simpler if all indices came in versions 1 and 2, and trackers were available for both.
Great article!
I really like how you focused on companies in the dividend ETF’s that reduced their dividends and got kicked out. That shows even the 25-year rule doesn’t guarantee a company won’t cut their dividends.
A good read would be Appendix C on how “Dividends are Not a Factor” in Your Complete Guide to Factor Investing by Berkin and Swedroe. They talk about how dividend policy is irrelevant to stock returns and how it’s never been challenged in the academic literature. Even Warren Buffet says he’s not going to offer a dividend because you can just sell the amount of stock equivalent to a dividend anyway.
Dividends do a have few advantages though, such as regular income without having to take any action, and you can avoid trying to figure out which investments to sell to maintain your allocation, but they don’t outweigh investing for total return.
Even if you look at the US markets, the iShares Dow Jones Dividend ETF underperformed the S&P500 ETF when comparing total return over the past 10 years. So in theory, you could have just bought the S&P500 and sold the equivalent numbers of shares regularly to collect your dividends, while still outperforming the dividend ETF.
@ROBH
Take a look at trustnet.com for more information on investment trusts. You can compare ETF to IT’s using their charting tool. I hold mostly IT’s in my portfolio and trustnet has helped me a lot. You’ll find the same thing though comparing equity-income IT’s to just equity IT’s – equity IT’s almost always outperform, so it’s best just to sell whatever you need for income rather than investing for income. One thing to watch out for though, especially for bond IT’s, is the wide bid-ask spread for thinly traded ones. That can often make it not worthwhile to buy for income.
@ FIRE in Thailand
Try using this great tool to help with maintaining asset allocations when buying or selling. I use it all of the time when putting lump sums into my SIPP to keep my allocations spot-on. Hope it helps!
http://optimalrebalancing.tk/?ckattempt=1&i=1
One of the problems with ETFs and unit trusts is that often it is difficult to obtain historical dividend performance beyond 5 or sometimes ten years. Investment trusts are public companies and therefore publish annual accounts which are available for download on the IT`s website. Most give dividend data going back at least 10 years – so accessing the annual report from 10 years ago will allow access to dividend data from 20 years back. It is interesting to look at their dividend payouts prior to and after the financial crisis of 2008 to see which income oriented trusts managed to maintain or increase payments during this difficult period. The majority the trusts in my income portfolio held up remarkably well even if capital values were badly hit. For me this is the true test of whether an IT should be in my portfolio.
I’m in agreement with the previous comment from “SEMIPASSIVE” a 2% yield is no good for me and too many of the so called dividend aristocrat shares may well have an unbroken dividend history but that’s not too difficult to achieve if the yield is only 2 or 3%.
@fire in Thailand
I think dividend policy can contribute to total return in that a company that struggles to fund productive use of retained profits, then a dividend policy of distribution of these earnings, allows the investor to reallocate these funds in a more profitable manner.
A dividend policy of distribution can avoid foolish takeovers, provides financial discipline.
Unfortunately since the crisis of 2008 and low interest rates then distortion occurs and as is often the way, Goodharts law comes into place, “when a messure becomes a target then it ceases to be a good measure”.
why did you leave out smart beta OEICs?
Timely article. As a soon-to-be-retiree doing “just one more year” I want to live off my investments, and also see their capital grow, somewhat contradictory wishes I will grant you.
I have a small Vanguard High Yield ETF holding which I am unlikely to expand as the overall performance is low. So I am keeping on with my alternate strategy, which is high yield bonds to cover expenses, and rely on the equity portion for increased capital (and often, decent yields).
FYI my bond portfolio is a bit like this https://www.fixedincomeinvestor.co.uk/x/analysis.html?type=model-portfolio&cat=analysis-comment&y=2015&aid=1348 and a bit like this https://www.fixedincomeinvestments.co.uk/fantasy-annuity-portfolio/
@Hariseldon
There are lots of reasons why dividends are a good idea for both companies and shareholders, but that still doesn’t make them a factor to consider when evaluating total returns. There are decades of research that confirm that.
Great post and some excellent comments !
So putting it all to the ultimate test. Say you had a lump some of £100k that you wanted to generate income from.
Would you do the above and split it into divi etfs? Or are there better alternatives for a relatively safe income stream – peer2peer, corporate bonds, reits etc etc ?
What you would do with that money is the ultimate test of the ETF strategy in this article !
If you invest in dividend orientated products it is inevitable that you will sacrifice long term growth as the underlying stock holdings will exclude high growth companies that pay pay zero or very low dividends and limit access to many international markets where companies are disincentivied from paying high dividends due to tax considerations. But investing for dividends should give you a high level of growing income – without the worry of having to sell stocks during market downturns which for many may be a big plus.
The sad fact of life is that’s there no free lunch. High growth implies high volatility. Reducing volatility through, for instance, bond holdings reduces growth. Anyone looking for retirement income will be well aware of these tradeoffs as whatever strategy you adopt to provide income has a cost in order to preserve capital and minimise the risk of income not keeping pace with inflation or dying before you do! This cost may be lower initial income, lower growth or the capital dying with you in the case of an annuity.