D.UN - Dream Office REIT

[KJP]

Dream Office REIT (TSE: D.UN; OTCBB: DRETF) is an externally managed REIT that owns office buildings across Canada.  It currently has a market cap of about $2.1 billion and an IFRS book value of about $3.5 billion.  I explain below why I think (i) there are good grounds to believe that Dream’s assets are worth roughly book value today; and (ii) management is properly incentivized to close the gap between market and book value and has a coherent plan to do so.

 

Value of Properties

 

Dream has 113.8 million units outstanding.  At $19/unit, Dream has a market cap of $2.1 billion.  Adding debt of about $3.4 billion yields an enterprise value of about $5.5 billion.  In return, investors get a portfolio of properties currently generating annual NOI of about $420 million plus about $250 million in other assets, including a $150 million stake in Dream Industrial REIT.  Subtracting $250 million from the EV of $5.5 billion results in an adjusted EV of about $5.25 billion for the property portfolio.  At an annual NOI of $420 million, the market is valuing the properties at about an 8% cap rate.

 

An 8% cap rate appears to be well above the private market value of similar buildings.  According to CBRE’s Q1 2016 Canadian market update, cap rates for Canadian office buildings currently range from mid-4’s to high-7’s, depending on building quality and location, with an average cap rate of around 6%.  Similarly, the cap rates Dream uses for its IFRS book value calculations largely mirror CBRE’s numbers, leading to management using a weighted average cap rate of about 6% for the portfolio as a whole. 

 

Unsurprisingly, cap rates have historically been correlated with interest rates.  I have no special insight into future movements of Canadian interest rates, Canadian office property values or the Canadian economy in general, nor have I done any investigation into Dream’s individual properties.  Instead, all I can say is that it appears that Dream’s IFRS book value appears to be much closer to the current private market value of Dream’s assets than Dream’s current enterprise value.

 

At the 6% cap rate used by management and suggested by CBRE’s data, the buildings would be worth about $6.7 billion.  Adding the other assets and subtracting debt yields an equity value of about $3.5 billion, or about $30/unit.  Dream also yields 8%.

 

If the value of the portfolio stays constant, and management manages to close half of the discount to book in two years, the unit price would increase to about $25, which is a CAGR of about 15%.  Add the 8% yield, and you’d have greater than 20% pre-tax returns over two years.  If half the gap is closed over three years, you’d have a high-teens annual pre-tax total return over that period.

 

If the entire market-to-book gap closes in two years, then you’d get a low 30’s annual pretax total return, and about a 25% annual pretax total return if the full gap closed in three years.

 

Management’s Plan to Close Gap

 

If management was properly incentivized and really believed in its book values, it could close the gap by selling properties at book and buying back shares.  That is exactly what management is doing.

 

Prior to April 2015, Dream Office’s external manager received an asset management fee based on the gross value of assets managed, which created no incentive to shrink the portfolio, even if doing so was in the best interests of unitholders.  In April 2015, Dream Office largely eliminated this conflict by extinguishing its obligation to pay asset management fees in exchange for 4.85 million REIT units.  As a result of this restructuring, the manager now has a very strong incentive to close the gap between Dream’s book value and its market value, and Dream no longer has the significant drag associated with paying the annual fees.  In addition, Dream (i) eliminated its DRIP, which was causing dilution through reinvestment of dividends at prices well below NAV and (ii) put in place a unit repurchase program. 

 

In February 2016, Dream Office formally announced a plan to sell $1.2 billion of properties over the next three years.  It already has sold (or has commitments to sell) about $500 million and has stated that the sales are occurring at book.  The proceeds are going to be used to delever and, hopefully, buy back shares. 

 

Risks

 

This investment is based on the apparent 200bps cushion between the cap rate the market is putting on Dream’s assets and their private market value.  That cushion would be eroded by rising cap rates, which could be caused by, among other things, rising interest rates or continued declines in the Canadian economy.  The risk is offset somewhat by the 8% yield, which will pay you something even if the market-to-book gap is closed by book value falling, rather than the unit price rising.   

 

[Picasso]

Cool idea. What are your thoughts on NOI growth?  Buyback math on REIT's is tricky because they tend to trade better on NOI growth rather than asset sales + buybacks. ACAS or CBL comes to mind. 

 

Also what do you make of remaining assets that won't be sold?  Will the operating results at the remaining properties be strong enough to continue to close the discount?

[KJP]

Cool idea. What are your thoughts on NOI growth?  Buyback math on REIT's is tricky because they tend to trade better on NOI growth rather than asset sales + buybacks. ACAS or CBL comes to mind. 

 

Also what do you make of remaining assets that won't be sold?  Will the operating results at the remaining properties be strong enough to continue to close the discount?

 

Here are links to two presentations on the asset sale plan:

 

http://www.dream.ca/office/wp-content/uploads/sites/4/2016/05/Dream-Office-Q4-2015-Strategic-Plan-Presentation-Feb-18-6-30pm-2.pdf

 

http://www.dream.ca/office/wp-content/uploads/sites/4/2016/05/D-UN-Q1-2016-Strategic-Plan-Update-v15.pdf

 

Based on the presentations and management commentary, they appear to be selling their middling assets and keeping both the crown jewels (high-end office buildings in CBDs, particularly Toronto) and the most troubled assets (primarily in Alberta).  It's presumably the Alberta assets that are driving the disconnect between private market value and NAV, so to some extent the sale plan is doubling down on management's belief that the unit price doesn't reflect the true value of these assets.  Balancing that out, however, is the fact that you're also increasing your concentration in the highest quality assets.   

 

I don't have any strong views on near-term NOI/unit.  Occupancy (~90%) and rental rates have been holding up (very slight increases per sqft) but management has said that Alberta will likely continue to require significant lease incentives, etc.  In addition, current rental rates are at or very slightly below market rates.  They are going to use some sale proceeds to delever (and have already done some delevering by paying off some high-cost debentures), so NOI/unit will likely decrease a bit in the short term.  Over the medium term, NOI/unit is going to depend on what management does with the sale proceeds and additional liquidity they've pulled together.  At the end of the day, if book value is solid and management uses a significant amount of the sale proceeds to buy back units at way below book value, they will eventually be able to raise the distribution per unit significantly, which will push up the price.  But I expect that's a few years away.

[KJP]

This is one way to shrink a big NAV discount:  http://www.dream.ca/office/wp-content/uploads/sites/4/2016/08/D-August-10-2016-financial-results-final.pdf