- Comments (12)
- Text Size: Down Up
moneysense.ca, 24/04/12
Estimating the Tax Hit on Dividends in Taxable Accounts
There is much enthusiasm among investors for dividend stocks these days. Some of the enthusiasm may even be warranted based on studies that have shown that dividend paying stocks have higher returns than non-dividend payers even though dividend payers have a lower standard deviation.
But there are also a lot of misconceptions about dividend payers. One widespread and persistent belief holds that Canadian dividend paying stocks should be held in taxable accounts to take “advantage” of the dividend tax credit. This belief is based on the fact that dividends receive the most favourable tax treatment for many tax brackets. For example, an Ontario resident with a taxable income of $50,000 will face a tax rate of 31% on interest income, 15.5% on capital gains and 13.4% on eligible dividends (Source: Ernst & Young 2012 Tax Calculator).
However, simply looking at marginal tax rates is a mistake. One also has to consider that dividend payments are made regularly and hence taxed at the hands of the investor each and every year whereas capital gains can be realized when the funds are needed, often after very long holding periods. To model the tax hit an investor incurs due to dividend payments, I constructed the spreadsheet available here. The spreadsheet has the following variables: the total return from stocks, the dividend yield, the tax rate on dividends and the tax rate on capital gains. The model assumes that capital gains are completely under the control of the investor. In reality, even the index fund with the lowest turnover will incur some capital gains distributions. Also, the rate of return and the dividend yield is assumed to be the same each year but as you well know, both these rates will fluctuate tremendously with changes in the level of stock prices. It is also assumed that the costs of assembling a portfolio of dividend payers is the same as that of tracking an index. As this is unlikely to be true in a real life portfolio, the model likely underestimates the effect of taxes on dividends.
If we assume that stocks return a total of 8% over a 25 year period, a $10,000 investment will grow into $68,484. Such an investment held in a taxable account will net an investor $59,420 (capital gains rate of 15.5%). As a base case, if we assume that the investment pays a dividend of 2.5%, it will grow into $57,431 (or about 2% less). Observe that even though the investor paid tax on dividends at a lower rate, they ended up with a smaller after-tax return on their original investment. If the dividend yield were 4%, the investment would have grown to $56,282. At 6%, it would grow to $54,798 and if all the stock returns were through dividends, the investment would grow to just $53,368.
It is an even worse decision for an Ontario resident with a taxable income of $100,000 to prefer dividend payers in her taxable accounts. In this case, the marginal tax rate on capital gains is 21.70% and 25.40% on eligible dividends. A $10,000 investment at 8% will grow to $55,793. If the dividend yield were 2.5%, the investment will only grow to $51,141. At a 5% dividend yield, the final amount drops to $46,990. Or look at it this way: a portfolio of stocks with a dividend yield of 5% must post total returns that is 0.40% higher than a portfolio of stocks with a dividend yield of 2.5% for an Ontario resident with a taxable income of $100,000 just to have the same after-tax returns.
moneysense.ca, 24/04/12









Hi,
Thanks for this; it is very thought provoking.
In your spreadsheet for the calculation of the “Dividend stock (after tax)” when you sell your shares isn’t the tax on the FMV – the ACB a capital gain (and not a dividend)? In you spreadsheet you are applying the dividend tax rate.
Also, if the FMV is less than the ACB then it would be a capital loss and I’m not sure how to represent that on the spreadsheet. I think your spreadsheet assumes that the stock return is greater than the dividend yield (otherwise the losses are not calculated correctly)….we don’t pay tax on losses.
Glen
I find this misleading. The calculation assumes that the f dividend stock is underperforming the market. In the default case the dividend stock has grown in value by 3% while the non-dividend stock grew by 8%. I believe that both stocks would often grow at about the same rate. Or is there data to suggest otherwise?
In this case, the dividend paying stock is a much more worthy investment a you reap both dividend income and capital gains.
Do you have access to any data on the relative performance odd these two categories of stocks?
Of course you knew that posting this would bring out those of us that enjoy watching the dividend payments roll into our accounts quarter after quarter.
In the case where a 5% dividend earner has to outperform a 2.5% dividend earner by 0.35% to generate the same after tax returns I would argue that it likely will beat that margin just by the difference in the dividend. I do not have any stats to back me up (and I will conceed that perhaps a stock with a 5% yield maybe has some other issues depending on the sector it is in) but I suspect that the capital gains in these two stocks will be comparable and that the 5% yielder wouldn’t lag the 2.5% yielder by more than 2.15% assuming the stocks are in the same sector and are close enough at a fundmental level that an investor would be indifferent to one company over the other.
It is interesting to hear the counter arguments to dividend investing. I think that focusing on total return is valuable. I also think that dividend investors should be aware of issues like the dividend tax rate being higher than the capital gains tax rate (at least in Canada for Canadian stocks) and the potential impact that paying dividends out to shareholders has on the ability of the company to grow their earnings. Also an investor interested in dividends in non-registered accounts needs to be aware of the year to year tax implications and needs to come up with strategies to deal with those tax implications that make sense for their situation.
@Glen: Thank you, thank you, thank you for catching my error. You are right, of course. I’ve corrected the spreadsheet and I will rework the numbers. I don’t expect that the conclusions will change, at least, I’m hoping it won’t or I’ll be eating a big serving of crow.
@Glen: Also, good point on the spreadsheet not handling dividend yield being greater than total returns. It is a limitation that I intended not to address but it would certainly be interesting to modify this to see where it leads us in that case.
@NLG: Not at all. The model assumes that the total return is the same from dividend payers as well as non-dividend payers. It then looks at the impact of taxes on after-tax returns. In other words, dividend paying stocks provide returns in the form of both dividends and capital gains. The dividends are assumed to be reinvested. The point of this model is to estimate the impact of taxes on dividend paying stocks in taxable accounts.
@0xcc: I don’t understand the comment: “In the case where a 5% dividend earner has to outperform a 2.5% dividend earner by 0.35% to generate the same after tax returns I would argue that it likely will beat that margin just by the difference in the dividend”. In both cases the total pre-tax return remains 8%. It’s just that the 2.5% yielder receives 2.5% in the form of dividends and 5.5% in the form of capital gains. The 5% yielder receives 5% in the form of dividends and 3% in the form of capital gains.
@CC – Sorry, I sort of changed parameters in my mind. My argument is that there wouldn’t be that much difference in the capital gains of two companies that an investor would be indifferent about choosign between and that the 5% yielder would make up that 0.35% gap but that isn’t your argument.
If I understand correctly your argument is that given a choice of return in the form of capital gains or dividends that an investor should choose capital gains in a non-registered account (and be indifferent in a registered account). I agree with that argument. Now if I could only find shares that would provide the dependability of dividends in the form of capital gains. Where did I put my crystal ball?
I had just started creating a spreadsheet almost exactly like yours for a post tomorrow. I’m glad I didn’t get too far with it before I read your post. You’re right that the value of deferring capital gains taxes is poorly understood.
[...] Capitalist said to be careful if you keep your dividend paying stocks only in a taxable account. Over time, I’m moving my dividend-payers to my TFSA. I certainly have a few payers [...]
Thank-you for a very thoughtful analysis. The analysis could be more realistic by assuming that given market conditions, the invester and/or fund manager will “rebalance” at least 2.5 to 6 % of the portfolio annually. Thus, the investor would ideally realize a capital gain return.
From a theoretical perspective, good post. Of course, in the real world history has shown that in general companies that don’t pay dividends, on average, don’t have very good return track records. In other words, you will have a difficult time finding companies that will have a total return in line with high-quality dividend paying companies. Therefore, if you invest in growth companies in your non-registered accounts and avoid the annual tax bill, you likely still have less money at the end of the day due to lower overall returns. Lastly, remember if you hold the dividend payers in your RRSP/RRIF the income will be taxed as income/interest when it comes out of the RRIF; putting you even further behind overall from a after-tax net worth standpoint. IMHO.
[...] Speaking of the Canadian Capitalist, they advised everyone how to go about Estimating the Tax Hit on Dividends in Taxable Accounts. [...]
Here in BC the dividend rate on the wife’s portfolio is negative due to wife’s marginal rate. If her income increases (which will probably not happen) we will slide some of the dividend payers into the TFSA. Corporate tax rates are falling and this anomaly will be reduced. Also like most dividend players we want a monthly cash flow, not a once in a lifetime vague chance at making capital gains.