For instance, if an ETF consisted of equal portions of the following stocks:
Stock ABC, dividend of 2%
Stock DEF, dividend of 2.5%
Stock GHI, dividend of 2.25%
Then you would expect the ETF holding those stocks to yield 2.25%. However, it would not. You much more likely would see a yield somewhere around 1.9%. He wanted to know why that was. So I did some digging and this is what I came up with:
1. The biggest difference is the fee the ETFs charge, while it is small (under 0.5%), the fee one charges makes a difference. They take that fee out of dividends to minimize tax implications to the ETF holders. For example, if SDY was supposed to yield 2.5%, but the expense ratio was .0.35%, the new yield is only 2.15%. By taking it out of the dividends received, it reduces taxable expenses, as opposed to liquidating positions.
2. The next is similar, as in it also relates to expenses, but about 0.1% (in one 0.15% in the other) reflects the transaction costs. Trading shares isn't free for an ETF either. They are constantly buying and selling shares to match the inflows and outflows of funds into the ETF. They don't eat these fees, they're passed on. In actuality this is about average as the transaction cost to most retail traders is higher or the same. For instance, if I bough 10 shares of Walmart (approximate $75, or a cost of 7500) the trade would cost (on average) about $9.00. Obviously if you can buy huge blocks of shares this goes down. These ETFs buy and sell LOTS of shares every day.
3. Then there is some bleed due to the fact that they are typically not 100% invested. In ETFs like SDY and VIG there are a LOT of cash inflows and outflows every day. Even if only 1% of the fund trades every day, they will have to be over 1% in cash. So if the ETF is only 98% invested on average, you receive 2% less dividends than if you were 100% invested. This lowers the yield too. As an example, if you were 100% invested and were going to receive a 2.5% dividend, by only being 98% invested, your dividend yield would drop to 2.45%.
Add all of that up, and you get a yield that "seems" anywhere from 0.3% - 0.5% to low. That said, MOST small and mid size retailer investors will lose that much on transaction and holding costs too, if not more.
Let's say I carry a theoretical $100,000.00 portfolio, and want to be very diverse and buy 50 different dividend stocks, bearing an average of 2.5% dividend (good luck). That means I allocate $2,000.00 to each stock. With the average dividend large cap stock price being around $40, that means I'm buying 50 shares, which on TDAmeritrade's retail platform would cost $7.99 a pop. So I lose 0.4% (.3995), acquiring the position. Assume I re-balance annually to make sure my position sizes stay proper, and I only have to make adjustments on 1/2 of my positions (more likely more, but lets say 1/2), then you lose another 0.2%. So my "effective" dividend is 1.9% -- or now worse than the ETF.
Obviously if you trade in large sizes, and hold longer without re-balancing, you can eventually beat the ETF dividend "bleed." But for most people that just isn't the case if you want the diversity, the balance, and the dividends.