In a recent comment (https://itawealth.com/huygens-portfolio-review-4-october-2021/#comment-31289) an ITA member asked us for our thoughts on an article entitled “8 Reasons Why I’m Not a Dividend Income Investor”.
Before addressing these “8 Reasons” I first want to re-iterate the point that we make all the time on this site – no single system will perform better than all others under all market conditions. Therefore, I believe that it is important to take the total cash that we have for investment and to divide it into sub-portfolios that offer diversification in terms of both the assets included in each “quiver” and the system we use to select and manage those assets. Within this framework I believe that “Dividend Income Investing” has its place. The allocation to this style of investing will vary from investor to investor for any number of individual/personal reasons – including some investors that may choose not to include a “Dividend Income” focused portion at all. In my younger days I was in this group and it is only as I’ve got older and into retirement that I’ve felt it prudent to include this “style” of investment in my portfolio holdings (probably should have done it long ago). At the other end of the spectrum there are probably investors that focus primarily, if not exclusively, on dividend income for wealth generation.
Having tried to put the question in perspective I’ll try to address the “8 Reasons”:
- Dividends result in a larger tax burden.
I am definitely not a tax expert so I can’t comment on specifics other than to say that, even if the fact that “you are invariably paying more in taxes than if you were holding non-dividend-paying assets” is true, it pre-supposes that the value of your “growth” assets is increasing – otherwise, better to pay taxes on something that is positive than not to have to pay taxes because you are losing money. If equities and bonds are inversely correlated and the dividends are coming from bonds then it is quite conceivable that it may still be advantageous to be holding bonds in a taxable account. An over-simplification, obviously, but I think it generally makes the point.
In Canada, we have a provision for “Dividend Tax Credits” that makes it difficult to generalize in terms of tax implications (it depends on an investors tax bracket and marginal tax rate) – but this is a jurisdictional complication so, to keep it simple I’ll just agree that, if focusing on dividend income this should be associated with a tax-advantaged account if possible.
- Dividends are not “free money.”
I would normally agree with this argument – and I often show examples of the reduction in an asset price as in goes ex-dividend. This is an accounting necessity/inevitability. However, there is evidence to suggest that the loss in value is often short-lived (soon forgotten by investors) and that a large portion of the loss is recovered quickly over the next few days or weeks. It is obviously difficult to quantify this since many other factors are affecting price action.
However, my approach to “Dividend Investment” is to focus on Closed-End-Funds (CEFs) rather than single stocks, Mutual Funds or ETFs. Because of the structure of these funds with a fixed, limited number of shares (as opposed to the varying number of shares held in Open-End Mutual Funds and ETFs) managed by specialists in specific areas of investment, there is more flexibility to alleviate the dividend offset effect. First of all, all dividends should be covered from Net Investment Income (NII) – i.e. the “income” generated by the manager from his choice of investment and how he/she manages it. This may include the use of leverage and derivatives to enhance “income” – and is pure alpha that is usually not available with other assets. Also, if there are withdrawals from Mutual Funds or ETFs the Fund manager may be obliged to sell assets held in the Fund at low prices. By contrast, a good manager of a CEF may be able to use his/her extra flexibility (use of leverage/derivatives) to avoid selling assets held in the Fund and thus reduce forced liquidations at low prices. This may have an impact on NII and may force the fund to reduce dividends – but this may be recoverable in time with significant “growth” consequences (outside of the original intent of income generation) – like Buy-And Hold without having to time a re-entry. Management fees are often high for CEFs (compared with MFs and ETFs) but good managers earn their keep. Provided that dividends are covered by NII and not subsidized by payments from the investor’s original Investment there should be little or no dilution. i.e. I should not be “getting paid with my own money”.
- Dividends limit total returns.
This argument assumes that if money is left in the investment that the investment will appreciate by “growth”. This may or may not be true (and will depend on the nature/structure/sector of the investment) – whereas the dividend is “guaranteed’ (at least to some degree – even if not at the same level). Total return = dividends + growth – if growth = 0 (or negative) we don’t have the possibility of any returns.
- Dividends are a forced withdrawal.
Maybe, but either this is ok because you need the money or you can reinvest (buy more shares) to improve geometric growth – essentially the same as if you had left money in a “growth” company. But companies with strong growth potential (that need to invest more money to grow) usually don’t pay dividends – so the asset class is somewhat different.
- I don’t want dividends.
Ok, so you’re a growth investor – focus on that – but don’t forget the power of geometric “income growth” that can come from dividends. Even with a long-term time horizon it just might be worthy of consideration for at least a portion of your portfolio. 8% “guaranteed” income, paid monthly over 20 years, can compound into a lot of dollars.
Stock buy-backs are good too – and have become more common in recent years – but not all growth companies do this and rarely on a regular basis (like monthly or quarterly dividends)
- Dividends only possess a psychological benefit.
This one I don’t agree with – they may certainly provide a psychological benefit, but it is not the only benefit. Do the math – 8% (annualized) dividend, paid monthly and reinvested – take a look at the geometric returns. Even without a “growth” component (i.e. only recovery of the original investment) do you believe that you can match that through growth alone ? (without getting a little lucky). What is the relative risk between the 2 investment styles? What are my drawdowns likely to look like?
- Dividend chasing decreases diversification
Again, I would disagree unless it is the only strategy employed – it could be argued that growth investing on it’s own also decreases diversification. Use both strategies for more diversification.
- Dividends are not guaranteed.
True – but neither is growth guaranteed. What happened in 2001-2002 and 2018? – even if the price of dividend paying assets dropped by the same (~50%) amount at least a portion of this may have been covered by dividends – and these were probably quite high (as a percentage) at the lows.
So, those are my thoughts on the article.
In summary, assemble a portfolio of sub-portfolios and include “Dividend Income” as the primary style target of one of those sub-portfolios. Hold this in a tax-advantaged account (probably a retirement account). My (personal) preference is to use CEFs rather than Mutual Funds or ETFs – these require a little more effort in terms of requiring due diligence but offer more diversification. MFs or ETFs are ok (and maybe a little less risky) if lower expected returns are acceptable. Don’t chase dividends without due diligence regarding whether dividends can be covered from NII – otherwise, the author of the article is correct, and you are not really earning the dividend – a portion of the distribution will just be a payback of your own investment. If you are using CEFs look for funds trading at a discount to NAV as a further opportunity to reduce risk.
The same ITA member that asked this question (Phil C) also asked me a couple of other questions off-line.
At the time I was in the process of reviewing my own portfolios and approaches to assembling a diversified set of sub-portfolios and to the relative sizes of these sub-portfolios.. In particular, I was giving a lot of thought to “core” portfolios (all assets held in the market at all times) and the pros and cons of risk parity. Long- time readers of this site may be aware that I have not been a big fan of risk parity – not because I don’t appreciate the benefits of using the technique to control risk – but because expected returns are significantly reduced due to the necessarily high allocations to low volatility (low correlation) assets required for diversification. I have been looking at leverage (and have introduced the Darwin Portfolio with 25% leverage) and have been considering the use of futures for cost efficiency in the application of leverage. At some point I hope to be able to share my findings. In the meantime, Phil has pointed me in the direction of new ETFs that are built around these principles so, at some point, I will be taking a deeper look at these. I am also encouraged by the fact that two “Quants” that I have a lot of respect for (Adam Butler from Resolve Management) – my inspiration for Adaptive Asset Allocation (momentum/rotation) – and Corey Hoffstein (Newfound Research) – my inspiration for the benefits of Tranching – are both looking at these new ETFs for portfolio construction.
I have also been asking myself a very simple question – what is Risk? In the financial world we get used to academics defining risk in terms of variance or Standard Deviation and this results in systems that sound imposing and often require a lot of explanation – but this is only a convenient way of quantifying “risk” for purposes of comparison – it isn’t a direct measurement of “risk”. It’s “the degree of wiggliness” in a temporal price series. Maybe something like “Loss of Value” is a simpler definition of risk – but how would one measure this in a meaningful way? Drawdown doesn’t really do it because it tells us nothing about the frequency of the drawdowns and the length of time for recovery that are surely important aspects of “risk”. Ok – so I’m coming back to earth and looking for a KISS answer – I’ll let you know if I find one – or you can offer an answer :).