Friday, 26 July 2013

Solid Canadian Stocks Currently at a Reasonable Price

When trying to pick stocks worth buying at current prices, we follow a different route from the riskier high volatility, high payoff stocks that are evidently the standard for the Superstars whose generally disappointing performance we reviewed last week.

Instead, we look for dependable performers in the mainstream TSX Composite index, combining stocks with:

A) traditional measures of value, such as
- a Price to Earnings ratio, lower than the overall market, currently 15.1 for the TSX
- high and consistent profitability, seen in Return On Equity (ROE)
- growing dividends in excess of the market average, that can be sustained as shown by Payout Ratio (dividends as percentage of earnings)
- we supplement this with BMO InvestorLine's automated Recognia assessment tool (available only to BMOIL clients) based on value investing principles to see whether it says the stock is under-valued, fair value or over-valued.
along with,

B) less commonly used measures that research suggests lead to better performance, such as
- presence of women on the board of directors, (data collected from Sedar's search tool by downloading each company's latest Proxy Circular)
- low dispersion of analysts' future Earnings Per Share estimates, (compiled from a combination of Yahoo Finance Canada e.g. here for Royal Bank of Canada, and BMO InvestorLine client-only stock research)
- low volatility of market price, (extracted by entering stock symbols in Globe Investor's My WatchList tool)
- companies with good ratings for social & environmental responsibility (SRI/ESG) if possible, though not all in our results table achieve that (based on membership in iShares Jantzi Social Index Fund - symbol XEN)
- governance ratings in the top half of companies , which means a rating of 69/100 and above (data from Globe and Mail Corporate Governance Board Games 2012)
- insider activity (data from TD Waterhouse client only stock research)

The Results
Fourteen stocks met our criteria. The banks dominate the comparison table below with six out of fourteen spots, with a scattering amongst other sectors. No utilities made the grade, due to their high P/Es. The mining sector is absent too, due to falling commodity prices and earnings that create high price volatility.

The surprising result is that almost all of the picks have seen insider executives and directors mainly selling shares in the past year. Whether that is merely cashing in to diversify wealth or to spend (several of the banks look like that), as opposed to a bet on the company's stock value, surely differs from one company to another. Imperial Oil is the only company with positives in every category. Even then, success is not 100% assured since, as we all know and should remember, the future may not be like the past and a highly successful company may begin to falter due to its own actions or simply changes to competitive or economic conditions. It is easily possible to find warning commentary about banks, energy companies, REITs, telecoms etc. Nevertheless we feel that on average these stocks should perform better than the overall TSX.

Disclaimer: this post is my opinion only and should not be construed as investment advice. Readers should be aware that the above comparisons are not an investment recommendation. They rest on other sources, whose accuracy is not guaranteed and the article may not interpret such results correctly. Do your homework before making any decisions and consider consulting a professional advisor.

Friday, 19 July 2013

Canadian Superstar Investors - How good is their performance?

A few years ago, the book Stock Market Superstars captured the thoughts of a dozen of Canada's biggest names in stock investing. A book review at the time noted that the 2008 financial crisis had severely knocked down all of a selection of funds managed by the Superstars. With drops of 41 to 57%, the Superstar funds fared much worse than the 33% fall of the TSX Composite. However, one bad year needs to be taken in a longer term context. How have these funds recovered and performed since then? Let's take a look. The table below shows the performance results we have compiled primarily using Morningstar Canada with supplementary data from the individual fund websites. All returns are net of fees.

One big winner
John Thiessen at the moment is the only investor who is still indubitably performing like a superstar. His Vertex Fund follows a very complex investing strategy, including considerable short positions and it seems to work. Vertex has bounced back strongly and is again handily beating benchmarks. The 2008 crisis looks like a temporary though gigantic dip in a steady upward climb.

Three "OK" mixed results
Three funds have achieved above-benchmark performance over either the latest five or ten years:
Eight significantly under-performing
The rest are falling well short of benchmarks over the past five years, seemingly not able to recover from the huge decline of late 2008 to early 2009. In several cases, where figures are available, the poor past five years have eroded tremendous out-performance up to 2008 to the point of now significantly under-performing benchmarks.

Very high volatility across the board
The superstars certainly have followed their philosophy of being different from the index, with the high numbers in the standard deviation column showing by how much more that has caused fund values to swing up and down. Hitching an investing ride with the superstars will be much wilder than a benchmark index and investors will tend to win big or lose big. The volatility can easily be seen in the following Morningstar price chart of the Resolute Fund, the most volatile and worst performing of all.

What to make of these results?
The first lesson is certainly that consistently beating an index over a long period is difficult, even for managers who have a reputation in the industry as being top guns. A second lesson might be that for many investors a simple and effective course would be to buy a broad market index fund such as the various ETFs we have reviewed in the past (e.g. in the case of Canadian equities, our reviews of large cap ETFs and low volatility vs large cap).

Disclaimer: this post is my opinion only and should not be construed as investment advice. Readers should be aware that the above comparisons are not an investment recommendation. They rest on other sources, whose accuracy is not guaranteed and the article may not interpret such results correctly. Do your homework before making any decisions and consider consulting a professional advisor.

Friday, 12 July 2013

What Happens to a Bond ETF When Interest Rates Rise?

From late May to mid June market interest rates took a relatively big and sudden jump upwards, anywhere from 0.5 to 1.0% across various bond maturities, as these Bank of Canada rate charts show. There was an immediate and appreciable effect on bonds and bond ETFs. After years of declining and ultra-low interest rates, the landscape may be changing for investors so it is opportune today to try looking forward and see how further rises might effect bond investments.

We'll focus on one ETF as a specific example - the iShares DEX Universe Bond Index Fund (TSX symbol: XBB) to work through the effects on several key elements of  bond fund performance:
  • Cash distributions
  • Share value
  • Return
  • Income breakdown by tax type of interest, capital gains and return of capital
Cash Distributions
Despite the recent increase in interest rates, the amount of cash being distributed every month by XBB is likely to continue dropping slowly, as it has done for the last ten years (see our post of June 28th with its graph of XBB's cash distributions per share). Until the return required from bonds goes up another 0.6% or so and stays at least that much higher for a considerable time, we believe several factors will combine to keep cash payouts on the decline.

XBB has enormous inertia
The first factor is sheer inertia due to the huge number of holdings in XBB - 735 according to its Overview tab. Change will take place slowly as the ETF's mandate is to passively track the DEX Universe Bond Index, which is the broadest index of investment grade bonds, and not to trade actively. Even though there is constant monthly rebalancing as newly issued bonds are added and some of the existing bonds exit, the monthly change is a small proportion of the total holdings.

New bond bond coupon rates match current interest rates
The second factor is that new bonds issued by governments and corporations will pay a coupon closely aligned to prevailing interest rates. In other words, bonds are issued as close as possible to $100 par value, give or take daily market movements around the actual day of issuance, and the coupon is about what investors expect as the return - e.g. see how succeeding  bond issues by the Government of Ontario for a seven year term to maturity had progressively lower coupons from 3.15% to 3.00% to 1.65% between 2010 and 2012 but they were always priced near $100.

Coupon rates of existing bonds in XBB are higher than required yield, therefore ...
Right now, on average, new bond coupon rates would still be below the vast majority of the bonds held by XBB. The return required by investors for XBB's bond portfolio is shown by the Weighted Average Yield to Maturity, which is 2.64% as of July 9th per the same Overview tab page. New bonds will, as an average, be entering XBB with coupons at about 2.62%. Meanwhile XBB's Weighted Average Coupon (same page) is 3.86%. Unless the bond market changes drastically and a disproportionate amount of long maturity bonds and/or from higher paying corporate issuers suddenly appear on the market, skewing entrants to XBB on the high coupon side, new bond issuance will still work to lower XBB's average coupon, and consequently its cash distributions. XBB is obliged to distribute all its interest income to share owners to avoid being taxed so the coupon interest drives the cash distribution level down as well.

Share Value
The immediate short-term effect of the rise in interest rates starting from the beginning of May has been a downward hit to the share price of XBB as the Google Finance chart below painfully shows.

That did not surprise most investors who know that there is an inverse relationship between bond price and interest rate - as interest rates go up, bond prices go down, or in the other direction, what has been happening for the last 30 years or so, as interest rates go down, bond prices go up. XBB is a very large collection of bonds so it reacts that way too.

The sensitivity of XBB's price to changes in interest rates is indicated by a variable called the Duration, which is also shown on XBB's Overview page. As of July 9th, the number is 6.80, meaning that for every 1% rise in interest rates XBB's price (i.e. the market value of the bonds in its portfolio) will fall about 6.8%. If we take a figure within the interest rate band we mentioned at the start of this post, say a 0.7% rise in interest rates, and multiply that by 6.8 we get a 4.2% drop, which is reasonably close to the price drop of 4.3% for XBB in the chart above. We note in passing that different bond funds will have differing Durations due to their holdings, as we found when we made compared fixed income alternatives in a previous post.

Premium bonds decline in value with time
All bonds mature at par ($100) no matter what price they were bought at. As we have already noted, XBB's portfolio consists almost entirely of bonds paying higher coupon rates than the required return /current interest rates..That means they are worth more and they sell for a higher price, which is why they are called premium bonds, i.e. priced above par.  But as time passes and they inexorably get closer to maturity, everyone knows that they will be redeemed at par at maturity. So the premium price declines steadily. XBB's premium bond portfolio will decline in value and price even if interest rates stay constant. Nevertheless, the investor is not doomed to inevitable net losses, since the higher coupon rates are still being paid out. The Yield to Maturity (YTM), also published on XBB's Overview page, tells us investors exactly what net return we can currently expect from XBB given the opposing effects of bond prices and interest received.

Rising interest rates will cause bond prices to fall and the premium prices to be eliminated. With XBB's YTM at 2.64% as of July 9th and the coupon rate at 3.87%, there is still more than a percent rise in interest rates to go before the premium bond effect disappears.

In figuring out what will happen to the net return from XBB, the place to start is Yield to Maturity (YTM). The rise in interest rates meant the overall market required a higher return from bond investments. The higher required return is reflected in a rise of XBB's YTM. At the end of March 2013, the YTM for XBB was 2.23% (per the iShares Fact Sheet) and had risen to 2.64% on July 9th (see also the YTM chart for the past year from the index provider).

But hold on, the observant reader may say, XBB just went down 4.2% in a month due to the interest rate rise and there would not have been enough interest cash paid out to compensate the big capital loss. How and when can an investor expect to make a positive return. How does YTM figure in?

Re-enter Duration as the "no net loss" time indicator
Duration has another meaning than sensitivity to interest rates. A calculation called Macaulay Duration (see Wikipedia details) gives a time in years at which the weighted average of the cash flows (interest and principal repayments) from a bond, or portfolio like XBB, are received. The significance for the investor of Macaulay Duration is that it will give a pretty close approximation of how long an investor needs to stay invested from the moment of buying XBB to be sure to make very close to the YTM as his/her return. Buy XBB today, hold on for the current 6.8 years duration, and the return will be about 2.6% - no matter what interest rates do till then, up or down! Whatever is lost on capital is made up by interest, or vice versa. Note that there are some sginificant assumptions that contribute to the accuracy of the approximation, such as all short to long term interest rates moving by the same amount when they change, and that yields of all bonds are equal (which is not really true as can be seen by listing the XBB holdings and comparing the individual bonds' yield to worst values). The calculation for sensitivity, called Modified Duration, differs slightly from Macaulay Duration but numbers are very close, such that it doesn't matter much for practical purposes of an individual investor in XBB that the iShares website's published Duration figure is in years (Macaulay) but the popup explanatory note talks of sensitivity (Modified).

MER taketh away return but rolling down the yield curve gives it back
In addition, XBB's management expense ratio would need to be subtracted from the YTM, though there is a good argument to be made that the constant pushing out of bonds reaching the one-year to maturity point serves to increase the yield somewhat (the so-called rolling down the yield curve strategy- see notes and links in this Finiki post) such that the effects offset each other.

Income Breakdown by Tax Type: Interest, Capital Gains and Return of CapitalInterest and Return of Capital are reasonably straightforward to foresee

To keep its non-taxable status, XBB must distribute to shareholders each year all the interest income it collects from the bonds in the portfolio. The coupon rate drives the interest income so interest distributions will follow a slow downward trajectory in the same way as we discussed above for cash distributions.

Return of Capital (ROC)
It is normal for XBB to throw off some ROC when new shares are created and the fund distributes cash according to the payout rate before the interest has really been generated by the new units. This would normally be quite small amounts unless XBB undergoes a huge inflow of investor money, such as happened in 2005 when XBB quadrupled in assets. In effect interest income is transformed into tax-deferred ROC. See also Rob Carrick's discussion of ROC in this Globe and Mail article.

Capital Gains
It is harder to predict exactly how much capital gains will be distributed in future though rising interest rates will operate to decrease gains, even up to the point of elimination if rates rise enough. XBB generates capital gains when its index tracking procedure forces it to sell bonds and those bonds are worth more than when they were purchased. We noted above that premium bonds decline steadily in value towards par as they move towards maturity. That tends constantly to reduce gains.

Bond rollover in XBB at one year to maturity won't generate gains
A predictable part of the turnover of XBB's portfolio is when bonds reach the point of one year to maturity and they are sold since they are considered at that point to be a money market security. Sorting the XBB holdings by maturity on the iShares website allows us to see that only 3 of 65 bonds that will get to within one year of maturity during the next twelve months have a coupon rate below its yield. In other words, these 62 are all premium bonds which will have declined in value since purchase from the natural decline towards par and from the recent rise in interest rates. There will be very little capital gains realized from selling those bonds.

Not much unrealized gains in the portfolio
Another factor is that when we look at the 2012 annual report for XBB, we see that as of 31st December 2012, the fair / market value of XBB's portfolio had only $86 million in unrealized gains on its bonds, or only 4% of the total asset base. Considering that interest rates have risen since then, the unrealized gains would have considerably diminished, perhaps even disappeared. If interest rates continue to rise in future, the unrealized gains surely will disappear and XBB will distribute none. Instead, XBB is likely to start generating capital losses, which are not distributed to shareholders.

What to do about rising rates
We investors have two choices - either hold XBB at least 7 years or so till its Duration, or buy a shorter Duration fund and accept the lower return that comes with it.

Disclaimer: this post is my opinion only and should not be construed as investment advice. Readers should be aware that the above comparisons are not an investment recommendation. They rest on other sources, whose accuracy is not guaranteed and the article may not interpret such results correctly. Do your homework before making any decisions and consider consulting a professional advisor.

Friday, 5 July 2013

Dividend ETFs and Stocks - Attractive Cash Distributions and Returns

Readers of this blog may remember from our January post The Crucial Difference between Price and Income Stability of Equities that an investment made at the end of 2006 would be receiving a higher cash payout from the iShares Dow Jones Canada Select Dividend Index Fund ETF (TSX symbol: XDV) than the the same amount invested in the iShares DEX Universe Bond Index Fund (XBB). XDV's distributions have grown strongly, also lengthening its cash payout lead over the mainstream equity ETF iShares S&P/TSX 60 Index Fund (XIU). Dividend funds are meant to pay out a lot of cash, so perhaps that is not too surprising.

However, the situation raises several questions since it is not a very long track record and is only for one dividend fund (see other Canadian dividend ETFs reviewed here and find a list of US-traded dividend ETFs in IndexUniverse's ETF Finder here). How stable and sustainable are the cash payouts likely to be? Does a dividend strategy sacrifice capital growth to achieve the cash payout, in other words how do dividend fund total returns compare to overall equity returns in the long run?

Cash payouts likely to be quite stable and growing - Dividends are "sticky"
Once a company has started paying dividends, management is loath to cut them except under dire circumstances. Thornburg Investment Management's Cultivating the Growth of the Dividend examines the long term history of dividend payments, both their stability and growth, in the USA. As the document discusses, management is very reluctant to cut a dividend as that conveys to investors that there are serious problems. On the other hand, paying a dividend limits the amount of profits sitting in the bank that managers might be tempted to use for wasteful pet projects or poor acquisitions and unprofitable empire building. Managers are forced to pick only the best projects that will add value. It is no accident that all sixty of the most successful and largest companies in Canada, as held within XIU, pay a dividend. About a sixth of companies in the TSX Composite do not pay any dividend as of July 2013 according to Norm Rothery writing in the Globe and Mail.

Dividends have always, and in countries around the world, many even more so than Canada and the USA, made up a significant portion of stock market returns. A chart from S&P Dow Jones in The Role of Dividends in Income Portfolios shows the significant contribution of dividends - anywhere from 25 to 40% - to the TSX Composite's total returns, decade by decade since the 1960s, with most of the variation being due to the variation in capital gains not the dividends!

Here is a chart from Thornburg's paper, showing the importance and stability of dividends in the USA going back to the 1870s; it clearly brings out the huge variation in capital gains / price returns.

Dividend paying companies produce higher total returns than non-payers
Yes, it is like having your cake (capital gains aka price return) and eating it too (dividends). Here is a chart from Mackenzie Investments showing results since 1989 amongst TSX stocks.
... and another showing what happened in the USA from Tweedy, Browne Company's The High Dividend Yield Return Advantage.

This latter graph, looked at carefully, introduces a layer of subtlety about dividends, namely which sub-segment of dividend payers does best - is it high dividend yielders, or low payout companies (dividends as a percent of profits), cash-flow (e.g. PH&N's Dividends: the Road Less Shaken) or some combination that gives the best results? Research result vary between countries, where different dividend cultures prevail. In the USA for example, there was a gradual shift away from companies giving back cash to shareholders through dividends. Instead, the companies did stock buybacks. Now the trend may be reversing. A recent market commentary by Scotia McLeod noted that 406 of the 500 companies in the S&P 500 now pay a dividend, the highest level since 1998. Reading through the excellent summary of the research in the Tweedy, Browne paper, it appears that a combination of high-yield and low-payout works best ... on average and in the longer term (a decade or more).But since the research on which exact dividend strategy works best differs, it is therefore no surprise to see dividend ETFs based on alternative selection and weighting criteria.

Dividend stocks and ETFs perform better in terms of return vs volatility
One attractive result, revealed quite consistently across many studies, is that dividends stocks have a higher Sharpe ratio, which measures the amount of return per unit of risk (standard deviation of stock price changes). Here is a typical chart for the TSX from Franklin Templeton. Similar charts can be found for the S&P 500 or other countries (e.g. Thornburg's Investing in Retirement Using a Global Dividend Strategy).

The Canadian dividend ETFs seem to be following that pattern within a very short time - according to BMO InvestorLine's ETF Compare tool, the trailing 3-year Sharpe ratio for the iShares S&P / TSX Capped Composite ETF (XIC) is 0.10 while the three dividend ETFs that have existed that long have much higher (=better) values: XDV at 0.30, CDZ at 0.38 and HAL at 0.28.The above-linked S&P Indices document shows that over the last ten years the two indices which underlie XDV and CDZ had both lower volatility and higher returns than the TSX Composite. XDV's index total returns were only marginally ahead of the TSX while the CDZ index returned a fabulous 2.5% per year more. On the other hand the CDZ fund had a fairly substantial decline in cash distributions between 2009 and 2011 while XDV did not as the chart below shows.S&P suggest that CDZ's emphasis on companies with a managed dividend policy, as opposed to XDV's emphasis on high dividends, may pay off better in the long run with a larger total of dividends and capital gains.

The outperformance of dividend strategies occurs more reliably over longer periods. There have been periods of several years when the market average does better than dividend stocks or funds. e.g. The above-linked S&P Indices document has a chart showing the TSX Composite outperforming the indices of both XDV's and CDZ's during 2006 to 2008 and the XDV index has been lagging the TSX for most of the time since 2005 to 2012.

What may be true of dividend stocks as a whole may not apply to every individual stock.  Some no-dividend growth companies will do extremely well  while some once-solid dividend companies decline and fall.

Bottom Line: Dividend stocks and funds offer good promise for solid, dependable, growing cash income as well as capital growth for long term investors.

Disclaimer: this post is my opinion only and should not be construed as investment advice. Readers should be aware that the above comparisons are not an investment recommendation. They rest on other sources, whose accuracy is not guaranteed and the article may not interpret such results correctly. Do your homework before making any decisions and consider consulting a professional advisor.