There are many ways of generating income from a portfolio. We consider how inflation and volatility influence two popular choices.

Recently we were asked to compare and contrast two different approaches to generating income in retirement:

  1. Buy Canadian dividend stocks and live off the dividend income.
  2. Buy a balanced portfolio of stocks and bonds and live off the total investment returns (capital gains, dividends and interest).

The journey has a few twists and turns. To begin we need to make some assumptions:

  • Our proxy for a dividend portfolio was XDV, the iShares Canadian Dividend ETF. The main points to know about XDV are that it tracks 30 Canadian paying dividend stocks selected on the basis of dividend growth, yield and payout ratio. This makes it a good proxy for investors who like to build their own portfolio of dividend stocks. The dividend yield over the past year was 4.89%.
  • Our proxy for a balanced portfolio was a mix of 4 ETFs representing Canadianinvestment grade bonds, Canadian stocks, U.S. stocks and International stocks. The portfolio has a total allocation 60% stocks, 40% bonds and is rebalanced annually.
  • The retirement period is 30 years. The initial portfolio value is $1 million and our retirees seek an income of $50,000 indexed to inflation.

We use data from 1st Jan 2006 to 31st Jan 20192 for both XDV and the balanced portfolio. The average annualised return for XDV was 5.34%, compared to 5.05% for the balanced portfolio. Compounded over 30 years $1,000 invested would be worth $4,762 if invested in XDV and $4,384 if invested in the balanced portfolio – a small win for XDV, but XDV is 100% stocks compared with only 60% stocks in the balanced portfolio. By comparison, an S&P/TSX Composite (an index of all Canadian stocks, not just dividend stocks) returned 5.31% annualised over the same period, not significantly different from dividend stocks.


Real Income for a real pension

Our retirees required their $50,000 withdrawal to increase with inflation. This is reasonable because retirees want to maintain their spending power in retirement. The impact of inflation can be a source of confusion. For example, an investor may say that they want to protect their principal. Usually they mean that they start with $1 million and end with $1 million. But $1 million 30 years in the future has a much-reduced spending power than $1 million today because inflation has eroded the purchasing power.  We assumed an annual inflation of 1.7% which is at the low end of historical inflation rates. Over 30 years the spending power of $1 million would be reduced to $603,063 in today’s dollars by inflation. Returns that include inflation are called nominal returns, returns that are adjusted for inflation are called real returns. To a good approximation, real returns equal nominal returns minus inflation. Our discussion will be in real returns (or today’s dollars).


Below is a plot of the portfolio value in today’s dollars with age for both the dividend and the balanced portfolios. The annual return is a constant 5.34% for the dividend portfolio 5.05% for the balanced portfolio. $50,000, indexed to inflation at 1.7%, is withdrawn at the start of every year.

Source: PWL Capital


Both the dividend and the balanced portfolio have declined as the $50,000 annual withdrawal, indexed for inflation, outpaces the gain from the investment returns. For most retirees, maximising retirement income without running out of money takes precedence over maintaining either the nominal or real value of the principal.


The rule is, jam tomorrow and jam yesterday- but never jam today

In addition to inflation, we know that market returns are not constant but vary considerably from year to year. Like the jam offered to Alice by the White Queen, volatility can be the promise of higher returns that never seem to arrive when you need them3.Volatility impacts how investments perform when subject to withdrawals. It is not hard to see why – when a market decline and withdrawal of funds happen at the same time, a recovery to previous levels becomes unlikely. To assess the impact of this risk, it is common to run many simulations that replicates the random sequence of returns in the market and then aggregate the results. We did this for the dividend and balanced portfolios using the historical volatilities for each. The results are shown below.

Source: PWL Capital


Now we have a more complicated picture. The solid lines represent the balanced portfolio and the dashed lines the dividend portfolio. In 10% of cases, markets will be good enough for the portfolio value to be above the green lines. In 10% of cases, markets will be poor enough to fall below the blue  lines. The yellow lines represent the average (median) outcome.

Key points:

  • Because of the volatility of stock market returns, both strategies risk running out of money. For both strategies, nearly 50% of outcomes result in a shortfall before age 95.
  • The risk of running out of money prematurely is greatest with the dividend strategy.
  • The dividend strategy has more upside than the balanced portfolio.

The dividend strategy has a higher probability of failure because it has a 40% higher volatility compared with a balanced portfolio. This may come as a surprise to anyone used to the idea that dividend stocks provide a reliable source of income while preserving capital. Make no mistake: dividend stocks are stocks, they have the volatility of stocks, and no amount of magical thinking4 will transform them into bonds.

How to weigh up the desire for high returns with the perils of the accompanying volatility? We suggest that retirees should take special note of the Sharpe ratio when comparing different retirement strategies. The Sharpe ratio is defined as the portfolio return minus a risk- free return, divided by the volatility. A higher Shape ratio indicates a higher return per unit of risk: it won’t tell you if you will run out of money, but it will suggest which strategy puts you most at risk.

The Sharpe ratio for the dividend portfolio is 0.39 and 0.565 for the balanced portfolio, an increase of more than 40%.

While planning with clients, we typically reject a strategy with more than a 10% risk of failure and neither the dividend nor the balanced portfolio meets that criteria. A lower withdrawal is one option, but as we have discussed elsewhere, using a flexible spending rule also reduces the risk of failure.


1 40% XBB, 20% XIC,20% XSP,20%XIN.
2 XDV was launched on 19/12/2005. Thanks to Raymond Kerzerho at PWL Capital for providing the historic data.
3 Lewis Caroll, Through the Looking Glass
4 You may think it is also easy to outperform a dividend ETF by picking your own stocks. The evidence of the past 10 years shows that no fund manager achieved this feat:
5 We have focused on historic data. Looking forward, expected returns and volatilities are both expected to be lower, without much impact on Sharpe ratios.