Beating An Index Through Sector Balanced Portfolio Construction

Summary

• There are arguably 3 main ways to generate alpha (1.Timing, 2.Sector Allocation, 3.Security Allocation).
• This article discusses how to generate alpha via #3 security allocation and an ETF, while focusing on the REIT sector.
• I chose the REIT sector because of its historic over-performance but recent under-performance. I believe the sector remains attractive.

Please read: This article was rejected by editors on the basis that it is too geared to the beginner investor. I am a little surprised to say the least since this is a method I have have only seen on the institutional buy-side, adapted for the "everyman" portfolio. It is one of 3 articles I have written over the last 2 weeks that have been rejected for various reasons. I am publishing via instablog and actively looking for feedback. If you could please let me know your thoughts in the comments section, I would really appreciate it.

Preamble

The focus of this article is to show you how you how the average investor can emulate strategies used at big Wealth Management firms to boost their portfolio returns and generate alpha. The three ways to create alpha: 1) timing, 2) sector allocation, and 3) security allocation (expanded below).

This article focuses on #3. More specifically: 1) how to find a suitable proxy for a Sector Index that trades on the open market and 2) how to create a position which will beat said Index using a mix of ETFs and individual securities.

The portfolio strategy in this article can be emulated on other sectors, but I focus the discussion on Canadian REITs.

As such, the Index of choice is the MSCI Canada IMI REIT Index and my proxy ETF is FTSE Canadian Capped REIT Index (TSE:VRE). I justify this proxy in the section on "Finding a Suitable Proxy."

How To Generate Alpha

As mentioned in the preamble, alpha is generated through three methods:

1. Timing
2. Sector Allocation
3. Security Allocation

1. Timing means you have the ability to get in and out of the market at the precise times to avoid draw-downs and ride upswings. Such a strategy depends on quick entries and exits and an edge to beat out all the traders, hedge funds, and algos competing for that same pie.

2. A Sector Allocation strategy focuses on over and under exposure to various sectors depending on where we are in the credit cycle. For example, late stage credit cycle normally means increased exposure to bonds, utilities, staples, gold, and REITs (maybe not in 2009). Early cycle, you’re looking at financials, technology, industrials, etc.

3. The focus of this article, Security Allocation, means you target a composite benchmark, let us say S&P/TSX, separate by GICS sub-sector and then look at all the equities that comprise that GICS sub-sector. You analyze each name and settle on a few stocks per industry that you like and think will outperform. Then you buy the names you like up to the point where your portfolio exposure in that sector is equivalent to that sector’s exposure in your benchmark index.

I like this strategy when working with larger sums of money ($100K+) to avoid transaction fees eating in your profit. It is a good way to target alpha while still keeping sight of the benchmark you are trying to beat.

An example of method 3 would be as follows:

1. Pick a total Index, say S&P/TSX Capped Composite (capped meaning constituents are capped at 10% of total index).
2. Verify sector breakdown, in this case Real Estate is 3.1%.
3. Find an appropriate ETF that mirrors the sector's performance.
4. Verify sector constituents (more on this in “Setting Your Sub-Portfolio Up for Alpha”).
5. Select equities in the sector that you expect to outperform (ie. CAR.UN as an example).
6. Purchase equities in #4 until they represent 3.1% of your portfolio.
7. Repeat with other sectors

How To Find a Suitable Proxy for Canadian REITS and MSCI Canada IMI REIT Index

(Finding a Candidate)

Unfortunately, you can’t buy the MSCI Index outright, nor would you if you are targeting alpha generation through security allocation. However, you can buy an ETF that mimics it closely.

I like Vanguard because of their lower fees generally, transparency, high AUM ~$5 Trillion, and attractive ownership structure.

Digging deeper we see the MSCI Index description is:

Source: https://www.msci.com/documents/10199/093c0f51-5d7c-4f05-bb30-a0051710c588

and (TSE:VRE) or FTSE Canadian Capped REIT Index description is:

Source: Our Products | Vanguard Canada Advisors

While VRE tracks the FTSE Canada All Cap Real Estate Capped 25% Index rather than the MSCI Index quoted above, I like that VRE has (in the description at least) the same exposure to large, mid, and small caps but I don’t like the higher MER of 0.39%.

MER is the first place where we can generate alpha by picking stocks, as long as we don’t overtrade, hold for extended periods, and have commissions eat into our returns. Keep in mind, you can buy VRE for very little commission which helps offset the cost of the MER on smaller trade sizes.

(Assessing Proxy Fit)

Looking at the Index differences we see 24 vs. 16 holdings between the MSCI and FTSE index respectively but similar P/Es and similar top 10 holdings.

Source: Our Products | Vanguard Canada Advisors

Source: https://www.msci.com/documents/10199/093c0f51-5d7c-4f05-bb30-a0051710c588

There is a difference in P/B or P/BV between the two funds but in general with mark to market now a big part of REIT financial statements and fair value gains/losses being pushed through the Income Statement and into Retained Earnings, a low P/B equity can potentially indicate troubled assets since fair value losses flow through the PnL.

In addition, based on the methodology used to select our alpha generators (see "Which Stocks to Buy"), ignoring low P/B equities is consistent with our selection methodology. So it is fine to use VRE as our proxy.

Setting Your Sub-Portfolio Up for Alpha

So, we have an ETF (TSE:VRE) which roughly mimics the characteristics of the MSCI index and is publicly listed. Using method 3 of alpha generation, if we wanted to construct a market beating sub-portfolio we have to look through VRE’s holdings and decide what has the potential to beat the Index.

Holdings are:

	VRE - May 31, 2020
Holding name	% of market value	Sector
Canadian Apartment Properties REIT(CAR.UN)	17.32689%	Residential REITs
Allied Properties REIT(AP.UN)	10.77642%	Industrial & Office REITs
RioCan REIT(REI.UN)	10.08479%	Retail REITs
FirstService Corp.(FSV)	8.59860%	Real Estate Services
Granite REIT(GRT.UN)	8.02493%	Industrial & Office REITs
Choice Properties REIT(CHP.UN)	7.24406%	Diversified REITs
H&R REIT(HR.UN)	6.07325%	Industrial & Office REITs
SmartCentres REIT(SRU.UN)	5.60117%	Retail REITs
Colliers International Group Inc.(CIGI)	5.40920%	Real Estate Services
Chartwell Retirement Residences(CSH.UN)	3.83601%	Real Estate Holding & Development
First Capital REIT(FCR.UN)	3.81571%	Real Estate Holding & Development
Northview Apartment REIT(NVU.UN)	3.61603%	Residential REITs
Cominar REIT(CUF.UN)	3.06653%	Diversified REITs
Boardwalk REIT(BEI.UN)	2.43024%	Residential REITs
Dream Office REIT(D.UN)	2.01973%	Industrial & Office REITs
Artis REIT(AX.UN)	1.95058%	Industrial & Office REITs

Source: Our Products | Vanguard Canada Advisors

Of course, cost is a concern so the smaller your pool of funds, the less holdings you get to pick else trading fees will eat at your profit.

So, what is a quick way to identify your alpha targets?  

Which Stocks to Buy?

(Methodology)

Efficient Market Hypothesis essentially states that a share price reflects all information available and that alpha generation is impossible. While I don’t agree with EMH for the three alpha generating methods I mentioned above, we can use this rationale to determine which equities we think will outperform by looking at recent share price movements, as they technically encompass the average opinion of a stock among all market participants and thus represent a collective consensus.

Source: TradingView

While an extremely messy chart, you can see that over the last 3 years only a handful of stocks have beaten VRE (at -0.30% relative performance since 2016). Excluding names without the “.UN” designation (UN indicates an income trust), out-performers are:

• NVU.UN (91.51%)
• CAR.UN (91.33%)
• GRT.UN (89.13%)
• AP.UN (27.00%)
• CHP.UN (16.89%)
• D.UN (10.49%)

This methodology of selection also holds true if you believe in a momentum type approach where past share movements and riding the trend should result in future out-performance. 

(Forward Looking Assumptions)

You can take things one step further by incorporating some high level, current knowledge of COVID-19. Based on general consensus, the sectors with more unknowns and thus more risk, at this time, are in the Retail and Office sub-sectors due to things like online shopping and work from home trends, respectively.

In this case, we assume the Industrial sub-sector holds strong because it is part of the supply chain supporting online sales which have been strong (plenty of articles on this, I won’t rehash this here and it does not mean I am against brick and mortar retail REITs). When we strip out Diversified, Office, or Retail REITs we are left with the following:

• NVU.UN
• CAR.UN
• GRT.UN

Now a quick google search will show that NVU.UN has an agreement to be acquired by Starlight and KingSett so we should exclude it from the list. A buyout agreement will usually shoot shares upward since all acquisition proposals feature some sort of premium to share price at the time to entice existing share holders. CAR.UN and GRT.UN are the only names left in our focus at this time.

Now is this analysis perfect? Not even close, in fact far from. If you’re familiar with the REIT space you may be thinking “hey, where’s IIP.UN, SMU.UN, KMP.UN or the host of other REITs that are we haven’t even looked at?” and you would be right to ask this.

This is a quick analysis to illustrate sector sub-portfolio construction that can be done in 35 – 45 minutes. Depending on your comfort, capital level, and experience, this may be all you need.

What I Would Buy To Beat TSE:VRE

If I were to construct a sector sub-portfolio from this analysis, I would buy 50% VRE, 25% CAR.UN, and 25% GRT.UN. Why?

• 50% VRE hedges if I made a in mistake picking CAR.UN and GRT.UN
• 50% VRE also saves us ~0.18% on the MER since the other two stocks don't have an additional cost besides buying and selling
• 25% of CAR.UN and GRT.UN gives us opportunities for capital appreciation in excess of VRE

In addition, I would monitor this portfolio to make sure CAR.UN and GRT.UN stay ahead of VRE in terms of relative capital returns. Should returns for either CAR.UN or GRT.UN after my purchase drop below a 10% under-performance relative to VRE (hypothetically VRE +2%, CAR.UN -8%) my analysis is obviously broken and my picks are not beating the Index. I say 10% because that is the maximum under-performance between the individual stocks and VRE since 2015.

In such a scenario, selling the under-performing equity(ies) and buying the Index is the responsible move and that is what I would do.

Next Steps And Conclusion

The purpose of this article is to demonstrate how one can approach alpha generation by stock picking while using at ETF to cheaply hedge over/under risk for a given sector.

In that respect, would I wholeheartedly recommend this method of investment to everyone? No. This article focuses on how one can create a sub-portfolio that has a chance at alpha, demonstrating this approach with REITs and showing you some high level work you can do to ball-park a sub-portfolio exposure.