The Effects of Interest Rates on Canadian Preferreds

A consideration to take into account when reviewing preferreds is the fact that they are sensitive to changes in interest rates.  The reasoning behind this is due to the structure of preferreds, as many issuances pay a relatively high dividend based on a percentage of par in perpetuity.  Like bonds, preferreds generally exhibit a negative relationship to interest rate changes.  When there is an increase in interest rates, the present value of future dividend payments decreases, and thus, the price of a preferred share would be expected to fall.

How can one estimate how sensitive preferreds are to interest rates? One way is to look at the effective duration of the asset class.  Duration is a tool that estimates price sensitivity to changes in interest rates; more specifically, it is the approximate percentage change in price resulting from a 100 basis point change in interest rates.

Using the empirical method by regressing historical portfolio returns of preferreds (represented by the S&P/TSX Preferred Share Index) to changes in interest rates, we found that preferreds in Canada have a historical duration estimate of -1.7.  This means that if interest rates were to rise by 1%, preferred prices would be expected to fall by 1.7%.  On the contrary, common stocks (represented by the S&P/TSX Composite Index) have a positive relationship with interest rate changes, with the historical duration estimated to be 15.5.

Using these duration estimates, we can look at how well interest rate changes have predicted the returns of the S&P/TSX Preferred Share Index during the time periods below.

Period Interest Rate Change Expected Return based on Duration Actual Return
2013 +0.88% -1.50% -7.16%
May 2013 – May 2014 +0.15% -0.26% -5.07%
YTD 2014 -0.46% +0.78% +1.80%

Sources: S&P Dow Jones Indices, Bank of Canada.  Duration estimate using data from Dec 2004 – Dec 2013 using monthly returns.  Portfolios are regressed against the 10-year Bank of Canada benchmark yield.

From the table above, we are able to see that the long-term historical duration estimate correctly projected the direction of the period return;but in all three time periods, the actual return was of greater magnitude than the expected return.  What is the reasoning for this?  Other factors besides interest rates also affect preferred prices. Some of these factors include company performance, call provisions of the specific share class, and the required credit spread of the preferred asset class above risk-free assets.  In 2013, the anticipation of future hikes to the target overnight rates in the U.S. and Canada also put negative pressure on the prices of preferreds.

Looking at the performance of each preferred share type in 2013 using the S&P/TSX Preferred Share Index, fixed rate preferreds performed the worst and floating rate preferreds performed the best.  Fixed perpetual preferreds carry the highest interest rate risk (i.e. duration), given that they have no set maturity date.  So it is no surprise that when interest rates rose in 2013, fixed preferreds had the lowest average price return.  The table below breaks down the average return for each preferred type for 2013.

Preferred Type Average Return
Fixed -7.0%
Floating -6.1%
Rate-Reset -6.5%

Source: S&P Dow Jones Indices.  Data from Jan 2013 – Dec 2013.

A preferred investor that seeks to mitigate the effects of increasing interest rates could look at shorter duration securities, such as floating-rate or rate-reset preferreds.  In 2013, where the markets saw increased rates, the S&P/TSX Preferred Share Laddered Index which is composed solely of rate-reset preferreds, had a total return of +0.88%.  The laddered index outperformed the S&P/TSX Preferred Share Index (total return of -2.64%) by 352 basis points.

The posts on this blog are opinions, not advice. Please read our disclaimers.

Leave a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>