How we score funds
The Income Durability Score summarizes how well a fund has turned its yield into something you could actually keep. It is out of 100, and it is built from fixed, published rules: with the scales and definitions on this page you can work out any fund's score from its numbers.
The score was built for covered-call and other option-income ETFs. Dividend, preferred share, REIT and CLO ETFs, and the preferred shares of split share corporations, are scored by the same rules, with one difference: they hold no options, so downside capture doesn't apply and they are scored on the other measures. Split share Class A shares keep downside capture, measured against the broad market of what the corporation holds, because the preferred shares leverage them. Cash and high-interest savings ETFs are not scored, because their yield follows the Bank of Canada rate: a lower payout says nothing about the fund.
It describes past data. It says nothing about future returns and nothing about whether a fund suits you.
What goes in
Each measure is converted to 0–100 on a fixed scale: the value in the "0 points" column earns 0, the value in the "100 points" column earns 100, with a straight line between and no points beyond either end. The score is the weighted average of the points.
| Group | Measure | Weight | 0 points | 100 points | What it asks |
|---|---|---|---|---|---|
| Income | Covered yield | 20 | 0% | 15% | Is the yield being earned? A fund paying 15% while returning 6% gets credit for 6%. |
| Income | Payout cuts | 15 | 30% a year | 0% a year | How much has the regular distribution been cut, not counting the swings of a payout that varies month to month? |
| Income | Payout trend | 5 | -20% | 5% | Is the regular distribution growing or shrinking? |
| Capital | Price return | 20 | -20% | 10% | Is the unit price holding up, or is the payout eating the capital? |
| Risk | Downside deviation | 10 | 25% | 2% | How big are the losing months? |
| Risk | Worst drop (3 years) | 10 | -40% | -5% | How far did it fall from a peak over the last three years? |
| Risk | Downside capture | 5 | 125% | 25% | In months the same stocks without the options fell, how much of the fall did the fund take? |
| Costs | Fee | 8 | 2.50% | 0.25% | What does the fund cost each year, trading costs included? |
| Costs | Liquidity | 7 | $10,000 | $1,000,000log scale | How much trades each day, so you can get in and out at a fair price? |
Low correlation to the S&P 500 is shown on each fund but not scored: it only helps if you already own US stocks.
The Very high / High / Medium / Low / Very low liquidity label shown in tables uses the same measure as the Liquidity score: average dollars traded per day (Very high from $1,000,000, High from $500,000, Medium from $100,000, Low from $25,000). Dollars, not units, so a $10 fund and a $50 fund are comparable.
How each measure is calculated
- Covered yield
- Current yield, credited only up to the fund's annualized total return over the window. A fund paying 15% while earning 6% gets credit for 6%.
- Payout cuts
- How much of the regular distribution was cut away, per year of the window. The payout level is the median of the last six months' regular payments (six for a monthly payer), each taken as a yearly rate so that switching from monthly to twice-monthly or quarterly payments isn't a cut, while skipped payments are; so a fund whose payout swings with its option income isn't counted as cutting each time it dips, and neither is one odd payment. A cut is a fall in that level of more than 2% from its high, and it counts by its size: a 3% trim adds 3%, a 40% cut adds 40%. Cuts compound, so two 50% cuts count as 75%, the share of the payout actually gone. One-off specials are ignored.
- Payout trend
- Annualized change in the regular distribution, from the last payment before the window to the latest one. Growth above 5% a year earns no extra credit.
- Price return
- Annualized change in the unit price over the window, before distributions: whether the payout is eating into capital.
- Downside deviation
- Annualized size of the monthly losses in total return over the window, counting gains as zero. Unlike volatility, a fund doesn't look safer for giving up its rallies.
- Worst drop (3 years)
- Largest fall in total return from a month-end peak within the window (not the fund's full history). Total return, so paying out distributions doesn't count as a fall.
- Downside capture
- Average monthly total return in months the fund's own benchmark fell, as a share of the benchmark's average fall. The benchmark is what the fund holds without the options: its uncovered sibling or an ETF of its index (BMO Equal Weight Banks for a Canadian bank fund), or its one stock; the S&P 500 where there is no clear one. Total return, in Canadian dollars unless the fund is hedged. 100% means the calls gave no cushion; leverage pushes it above 100%. Not scored for dividend, preferred share, split share preferred, REIT and CLO funds, which write no calls. Split share Class A shares are scored against the broad market of what the corporation holds, since the preferred shares leverage them.
- Fee
- Yearly cost: the management expense ratio (MER) plus the trading expense ratio (TER), from the issuer, its latest management report or its ETF Facts. The TER is the commissions and other costs of trading the portfolio, which the MER leaves out and which add up for funds that write options actively. For funds of funds both include the costs of the ETFs they hold. For leveraged funds the interest on borrowing is in one or the other, depending on the issuer. Where no TER is published, the median TER of funds of the same type is used; where no MER is published yet (new funds), the management fee.
- Liquidity
- Average daily traded value in dollars over the last 63 trading days, on a log scale.
Rules every measure follows
- Returns are monthly, from month-end values, and the window ends at the last month-end, so a score only moves when a month ends or the yield, fee or liquidity changes. Total return includes every cash distribution, reinvested. Year-end distributions paid in new units are left out: the units are consolidated, so the price doesn't drop and nothing was paid.
- Total return, not price, for downside deviation, worst drop and downside capture, so paying out a distribution never counts as a loss. Only price return uses the unit price alone.
- Regular payments: a payment more than 1.5 times the median of the three before and after it is a special and is left out, and so is an extra paid within a week of the one before. Yield is the latest regular payment times the payments a year, over today's price.
- Short histories: a fund with fewer than 36 months is scored on a blend. Its own value counts for its months ÷ 36 (with 12 months, 33%). The rest is a below-average value for its type: the one 75% of funds of that type with a full 36 months do better than (all such funds, if its type has fewer than 5). Under 18 months, downside capture uses that value alone: the fund's own swings too much over a handful of down months.
- Who is scored: funds with at least 12 months of returns and data for at least 60% of the weight. A measure a fund has no data for is left out of its average, and its page says so.
- Ranks compare funds of the same type only. There is no overall rank: scores of different types partly reflect what their markets did.
Fairness rules
One window
Return, risk and payout measures use the same trailing window, up to 36 months to the last month-end, so no fund is judged on a different period and scores don't drift between month-ends.
Short histories
A fund needs at least 12 months to be scored. Under 36 months, its own record is blended with a below-average fund of its type, so an unproven fund has to post a good record to rank well.
Fixed scales
A fund's score changes only when its own data changes, not when other funds join or leave.
No caps for leverage or single-stock funds
Their extra risk shows up in the measures and is flagged on the page. Ranks compare funds of the same type only.
What the score doesn't tell you
- It depends on the last three years. Covered calls lag in strong rallies and look best in flat or choppy markets, so scores move as the window rolls forward, and funds of different types partly rank on what their markets did. Compare funds of the same type: each fund page gives its rank within its type.
- Downside capture rests on our choice of benchmark. Each fund is measured against what it holds without the options: its uncovered sibling ETF or an ETF of its index, which we pick from the issuer's description and holdings (each fund page names it), or its one stock. A fund that mixes sectors or countries has no clear equivalent and is measured against the S&P 500, which partly measures how different that mix is. Three years still hold only a dozen or so down months, so it carries a weight of 5.
- The fee is the MER plus the TER, where we have the TER. The trading expense ratio (TER) is the fund's commissions and other trading costs, which add up for funds that write options actively; some issuers also put the interest on leverage there rather than in the MER. Funds whose TER we don't have yet, mostly those whose figures we enter by hand, are scored with the median TER of funds of the same type, and their page says so. Neither figure includes the bid-ask spread you pay when you buy and sell.
- Liquidity is dollars traded. What you pay to trade depends more on the bid-ask spread, set by the holdings and the market maker; a thinly traded fund of big banks can still have a tight spread. Fund size isn't scored yet either; fund pages show net assets where the issuer publishes them.
- Tax isn't scored. How a distribution is taxed (return of capital, capital gains, Canadian dividends, foreign income) changes what you keep in a taxable account, and US withholding tax applies to US stocks even inside a TFSA. Fund pages show the latest year's breakdown from the issuer's annual tax summary, where it publishes one.
Fund versus its top 10 stocks
On each fund page we take the fund's latest published top 10 holdings, scale them to 100%, and ask what they would have done if you had bought them at the start of the period and held them, with dividends reinvested, in Canadian dollars. We compare that with the fund's own total return, income, and worst drop. Because the fund may not have held these stocks earlier, this is a rough yardstick for what the covered-call strategy and fees added or gave up, not a record of what you could have done. Periods longer than five years are not shown, because today's top 10 are mostly the winners.
A single-stock fund is compared with its one stock instead: buying the shares at the start of the period and holding them, with dividends reinvested, in Canadian dollars, without the borrowing many of these funds use to buy more shares. The fund has held that stock since launch, so every period back to launch is shown.
Funds that beat their benchmark ranks these comparisons over 1 and 3 years, and adds one more: a covered call fund against its benchmark, the uncovered ETF of the stocks it writes options on, as its own documents describe them, and a dividend, preferred share, REIT or split share fund against the broad market for its region (XIC for Canada, SPY for the US, and so on), which is our choice rather than the index the fund tracks. Both are measured the same way over the same days; a fund hedged to the Canadian dollar is measured in the other side's own currency. The full list shows each benchmark and market yardstick and the funds compared with it.
Where the data comes from
Prices and distributions come from public market data. Fees, leverage and holdings come from each fund provider's website. Data is refreshed regularly and may lag; each page shows its date. Errors happen. Please check the provider's prospectus and ETF Facts.