My Household Portfolio Management Process - Part 3: Asset Allocation

by: Michael Patenaude

Summary

Every investor has their own needs, objectives and risk profile.

Sharing portfolio management approaches can improve performance by sharing knowledge and experience.

My household portfolio plan is shared for comments and improvement.

Part three in this series focuses on portfolio asset allocation.

Introduction

This is the third in a four-part series on how I manage our household portfolio as my wife and I start to consider retiring from full-time employment in about five years. The first article was about our investment objectives and the second was about my risk management approach. If you've not read them yet, kindly do so and you will have the necessary context for this article.

As a reminder, I am trying to highlight some of the key aspects of portfolio management an investor can control, namely their objectives, risk management approach and asset allocation. These variables are all closely related to each other and thinking carefully about them can help the self-directed investor focus on some key decisions with regards to their portfolio management approach.

My Household Portfolio Management Process

Part 3 : Asset Allocation

If I had to characterize my asset allocation approach, I would refer to it as primarily an "offsetting strategy."

What I mean by "offsetting" is that I try to divide the assets in our portfolio into categories that hopefully balance each other to some degree. Failing that, I try to ensure we don't have too many assets in the same category.

There is nothing terribly new about this. For decades the classic offsetting allocations have been between stocks and bonds: 50/50, 60/40, 70/30 and 90/10, respectively, are typical splits between these two asset classes.

Other asset class variants have emerged, including Harry Browne's permanent portfolio, which I referred to in my first article of this series. The permanent portfolio, in addition to stocks and bonds, also brings cash and gold to the mix.

As mentioned previously, I don't follow the permanent portfolio verbatim; not even close. I do borrow and modify the asset categorization, though.

As part of my analysis to set up our portfolio asset allocations, I have run some back-of-the-napkin calculations (see the "stress test" below) on how our current portfolio might have fared during the 2007-2009 period. This analysis allowed me to refine the proportions I've allocated to each asset category so that I can sleep better (if not well) at night.

Equity Allocation

I, like many others, think about risk/return trade offs and my reading leads me to conclude (like most) that stocks offer the best risk/return potential (compared to bonds, cash and bullion-related assets).

As a result, I weight stocks the highest in my allocation at 45% of our portfolio (49% if I included bullion-related stocks). Many would consider that a conservative weighting, and I'm fine with that. I think I know myself well enough that if I have more than that in stocks, I will not sleep well at night if equities drop in value significantly.

So, 45% was a bit of a ballpark choice based on factors like 100 minus current age (which would suggest, in our cases, about 45%). Note I did not use the more "modern" 110 minus current age (which is meant to reflect longer lifespans) as that would put me closer to 55% in stocks. To me, 50% is my upper limit for one asset class, otherwise we are starting to make too big a bet on one category.

Bonds, Cash and Bullion-related Assets

Using my "offsetting strategy," I look to other major asset classes to balance out our stocks. Bonds are the obvious and traditional choice. I could have picked a 45% bond allocation to equally match the stocks, but, as noted, I take into account additional asset categories: cash and bullion-related.

So with 45% in stocks, I have 55% left to redistribute across bonds, cash and bullion-related assets. After a lot of thought, I have come up with 28.75% for bonds, 16.25% for cash and 10% for bullion-related.

Since the cash is 10% true cash and 6.25% bonds with less than one-year duration, I could also characterize my allocation as 35% bonds and 10% cash. These two combined total to 45%, an "offset" for our stocks at 45%.

A further note on our true cash. I aim to keep 50% of it in American dollars, and 50% of it in Canadian dollars as yet another "offset."

That leaves 10% to allocate. I chose bullion-related assets as I have been somewhat influenced by the permanent portfolio and I have been influenced by gold's behavior during and after the 2007-2009 market meltdown (and again more recently in the first quarter of 2016). I noted earlier that our bullion-related holdings include both actual bullion and mining stocks (60/40 respectively).

Real Estate

Note that I have not mentioned real estate, as we own two properties already, and I don't want to put too many more eggs in that basket (even though our real estate is not part of our investment portfolio).

I have nothing against real estate, and in fact, it has historically been our best performing asset class (we used to own a couple of rental properties).

Numerous reports have surfaced about how overvalued Canadian real estate is, so I am a bit cautious (but see Table 3 below, for more on this asset category, as I do make limited provision for it).

Target Asset Category Allocations - Asset Class

In summary, the target asset category allocations I use are found in Table 1.

Table 1: Target Asset Category Allocation

Asset Category

Target Percentage of Total Portfolio

Stocks

45.00%

Bonds

28.75%

Cash/near cash

16.25%

Bullion-related

10.00%

Total

100.00%

Click to enlarge

Within each category, I have further sub-allocations. For stocks, I have income, growth, mixed (typically international ETFs) and high risk; for bonds I only have "corporates" with a maximum five-year duration to maturity (given our current yield climate); for cash I include bonds with a duration to maturity of less than one year; and for bullion I have both true bullion and bullion equities.

Table 2: Further Breakdown of Target Asset Categories

Category

Sub-category

Target Percentage of Portfolio

Totals

Stocks

Income

22.50%

45.00%

Growth

16.25%

Mixed (ETFs)

3.75%

High risk

2.50%

Bonds

Corporate bonds with 1 to 5 year duration

28.75%

28.75%

Cash

True cash

10.00%

16.25%

Corporate bonds with less than 1 year duration

6.25%

Bullion

True bullion

6.00%

10.00%

Bullion stocks

4.00%

Total

100.00%

Click to enlarge

Target Asset Category Allocations - Sector

I further break down equities into industry sector (using a slightly modified version of typical sector breakdowns). I wish to have a presence in all sectors (since I can't predict which will outperform or underperform), but I tend to pair them and favor some over others.

The basic "offsetting" principle I follow is if I have something I perceive to be higher risk, it should be more or less equally paired by something I perceive to be lower risk.

I also consider the "offset" of growth vs. income stocks and try to keep growth and income at 50% each of our stock holdings.

So, for instance, the technology sector is generally higher risk and growth-oriented. By contrast, the financial sector is usually lower risk (at least in Canada) and income-oriented.

So I have an equal allocation to each. I follow that pattern across our stocks as illustrated in Table 3 (note that due to the sectors I use, I had to split consumer discretionary across both growth and income in "offset" #5).

Table 3: Target Sector Allocation of Equities

"Offset"

Sector

Goal: Higher risk, growth

Goal: Lower risk, income

1

Technology

15.00%

Finance

15.00%

2

Industrials

10.00%

Utilities

10.00%

3

Materials

10.00%

Telecom

10.00%

4

Healthcare

7.50%

Consumer Staples

7.50%

5

Energy

5.00%

Real Estate

5.00%

Consumer Discretionary

2.50%

2.50%

Total

50%

50%

Click to enlarge

Target Asset Category Allocations - Geographic Location

Finally, I also look at geographic distribution of our stocks only, and of our total portfolio.

My perspective is affected by the fact we are Canadians, living in Canada, and using most of our financial resources in Canada. I recognize that the Canadian stock market represents about 4% of the world's stock market capitalization, so should have some, if not considerable exposure outside Canada.

With such easy access to US stock exchanges through discount brokerages here in Canada, the obvious first choice for geographic diversification is to the US. In addition to the practical access, the US market is, of course, about 50% of the world's market value, and provides for literally thousands of investment products, stocks and other assets, many of which are not available in Canada.

Beyond the US, I also try to have a bit of exposure to the international developed and international emerging markets.

Obviously, one factor in the decision to allocate across geographic regions is currency. Since most of our holdings are in Canadian dollars, I put all our US and international holdings in US dollars (none are hedged). From a stock-only perspective, that means we have 65%/35%, respectively, in Canadian and US dollars. This is yet another dimension of offsetting holdings that I consider.

Here is the target stock-only breakdown of my geographic allocation:

Table 4: Target Geographic Allocation - Equities Only

Canada

US*

International - Developed*

International - Emerging*

65%

25%

7%

3%

Click to enlarge

*Held in US dollars.

Here is the target total portfolio breakdown of my geographic allocation (the main difference here is that half our cash is in US dollars in our trading accounts and all our bonds are in Canadian dollars):

Table 5: Target Geographic Allocation - Total Portfolio

Canada

US*

International - Developed*

International - Emerging*

77.5%

17.5%

3.5%

1.5%

Click to enlarge

*Held in US dollars.

Linking Portfolio Objectives and Risk Management to Asset Allocations

I don't think this article would be complete without showing how our portfolio objectives and risk tolerance is supported by these asset allocations.

Our Asset Allocations Support Our Portfolio Objectives

We wish to slow down our participation in the workforce starting in about five years, and have enough money in our portfolio to maintain our standard of living in retirement. Given the pretty conservative asset allocation we've chosen, there is a possibility our portfolio will underperform over time. This may force us to delay plans to retire or to change our portfolio allocation in hopes of getting better overall performance. This is on my list of considerations when I next review our household investment plan.

Our portfolio total return objective (without new contributions) is 7%. We therefore need to be focused on capital growth and income. 90% of our portfolio is either generating capital gains, income, or a combination of the two (10% is in true cash and therefore is not productive).

We do want to have enough cash available to avoid selling securities that are materially down in value. We can easily achieve this objective over the next five or six years since we're both still working, have cash on hand, and are still saving.

Along with our cash on hand and new savings, we also have a steady annual stream of maturing bonds providing additional cash. Maturing bonds represent about a year's income, another one of our objectives.

I would say that we have a good mix of cash, bonds and bullion-related assets for me to have the sleep well at night factor under control. Unless markets go south badly, if anything, I may begin losing sleep by not having enough performance built into our portfolio, and falling short of our 7% return objective.

I may have to revisit our cash allocation in the future (but see below on supporting our risk management objectives).

Our Asset Allocations Support Our Risk Management Approach

I believe we've not only structured our asset allocation to support our objectives (with the caveats noted above), we have also allocated capital to support our risk management approach.

Our main objective from a risk management approach is to avoid long-term capital impairment. While there are certainly no guarantees in life, our allocation to cash, bonds and precious metals gives me confidence we have addressed this risk factor.

We wish to avoid a total portfolio decline greater than 10% in any given year. I think, again, our asset allocation gives us pretty good down-side protection (see below for a bit more on this). I don't think anyone investing significantly in the stock market can be fully protected without extensive hedging. So I am trying to be reasonable with my risk management approach.

Our portfolio avoids excessive day-to-day volatility. It is a pretty volatile day if our total portfolio moves by more than 3/4 of one percent. That tells me the offsets I've put in place are largely working, even during times of somewhat elevated volatility like that of the first quarter of 2016.

We have combined growth and income stocks and are pretty equally balanced in both.

We generally avoid assets with less than investment grade quality. The prime exceptions would be the small portion of our portfolio invested in high-risk growth stocks. Our risk management approach does allow for a small percentage of our portfolio to be held in this type of asset.

We have avoided micro-caps and penny stocks. Our holding with the smallest market capitalization value is $468M CDN (about $350M US). We only have a handful of holdings below $1 billion in market capitalization.

My Back of the Napkin Stress Test for Our Portfolio

The next topic I'd like to explore in this article, however non-empirically, is how our asset allocation might hold up if we experienced another financial crisis such as that between 2007-09. Obviously, there is zero chance of history repeating itself exactly, but if a similar crisis were to take place, I'd like to "stress test" its impact on our portfolio.

The financial crisis, as defined by the behavior of the S&P 500 index from the peak in October 2007 to the trough in March 2009 and followed by a return to recovery in March 2013, was exemplified by:

  • A 56% drop in the S&P 500 to the trough, followed by a 130% rebound to recoup all losses.
  • A 47% drop in the S&P TSX Composite to the trough, followed by a 68% rebound.
  • A 62% drop in the MSCI EAFE (tracked by the iShares EFA ETF) to the trough, followed by an 81% rebound.
  • A 5% drop in Canadian bond values to the trough, followed by a 13% increase (as represented by PH&N D series high yield bond ETF).
  • A 25% increase in the price of gold, followed by a further 71% increase in price by the time the S&P 500 had recovered.
  • A 33% increase in the US dollar vs. Canadian dollar to the trough, followed by a 21% decline until the recovery (note this impacts all of our $US holdings and is factored into the stress test).

If our portfolio were subjected to the same crisis that occurred as described above, I estimate we'd face a 22% peak to trough drop before seeing a 5% gain, in Canadian dollar terms.

It is important to emphasize I have converted US denominated assets to Canadian dollars, using "crisis period" exchange rates, to estimate the impact of the "meltdown" on our representative portfolio in Canadian dollar terms. That explains why, for instance, the "recovery value" for the S&P 500 is less than the peak value (currency exchange rates account for the difference).

Please note I'm making a lot of simplifying assumptions and approximations here (really, I'm not kidding), including:

  • Using the PH&N Series D bond ETF to approximate our individual bond holdings;
  • Assuming stocks in our portfolio behave, proportionately like the S&P 500, TSX Composite and the iShares EFA ETF that tracks the MSCI EAFE;
  • Assuming identical US/Canadian dollar exchange rates for the time frames involved; and
  • Ignoring income from dividends and interest.

So, with these major caveats, Table 6 presents the results of the stress test.

Table 6 - Fictional Stress Test on a $100,000 Portfolio

Using Our Current Asset Allocations

($CDN)

Asset

Peak Value

(Oct 2007)

Trough Value

(Mar 2009)

Recovery Value

(Mar 2013)

Bonds (PH&N Series D ETF)

34,088

32,264

36,390

Canadian Stocks (TSX Composite)

32,936

17,490

29,374

US Stocks (held in $US account) (S&P 500)

13,505

7,853

10,726

International Stocks (held in $US account) (EFA ETF)

4,915

2,507

2,698

Gold

6,296

7,862

13,425

$CDN Cash

4,213

4,213

4,213

$US Cash

4,048

5,384

4,257

Total

100,000

77,572

104,562

Change ($CDN)

n/a

-22,428

4,562

Change (%)

n/a

-22.4%

4.6%

Click to enlarge

While not great performance from peak to trough, it does reassure me slightly that our asset allocation may be protecting us from over 60% of the downside of a major US stock market meltdown (and its knock-on effects around the globe), and, to see that once the US market has recovered, we're almost 5% ahead of it.

Again, I am fully aware that I am making numerous assumptions and approximations, and that history is not going to ever repeat itself. Yet, it is an instructive exercise to complete, if only to have a better appreciation of what assets might move up or down in a crisis situation.

Do We Have the Right Allocation to Achieve the Growth We Want?

The final topic I'd like to discuss is whether or not we have the right asset allocation to meet our upside objectives.

Looking at only columns two and three of Table 6 may give some clues about upside performance in our portfolio.

Between March 2009 and March 2013, the S&P 500 returned about 130% in US dollar terms.

Our total portfolio, while made up of much more than only US stocks, would have returned about 35%, in Canadian dollar terms.

Between March 2009 and March 2013, the TSX Composite rose about 68%. Again, our total mixed portfolio would have gained about 35%.

I realize that comparing our mixed portfolio in this manner is very crude, and is somewhat akin to comparing apples to oranges. Notwithstanding these limitations, it does provide some clues as to how we might make out in a strong stock market environment.

Of course, many other scenarios are possible. For instance, looking only at the first quarter of 2016, our portfolio is up 1.2%, compared to the TSX Composite which is up 3.67% and the S&P 500, up only 0.77%.

In the most recent quarter, the Canadian index is outperforming the US one, in large part due to a heavy concentration in energy, materials and financial services. These sectors have done reasonably well to date this year. Our portfolio is not keeping up with the TSX Composite, nor would it be likely to if its main constituents continue to outpace.

However, unlike in the recovery from the trough of the financial crisis, our portfolio is currently outperforming the US index. That would appear to be due to our exposure to materials (bullion) and the general weakness in the US index, which is a proxy for only about 18% of our total portfolio.

Summary

In this article, I have laid out my approach for "offsetting" asset allocations. I believe that by using "offsets," or balancing one category of assets against another, I am better able to diversify our holdings and meet our risk management objectives.

I have done a "back-of-the-napkin" calculation to "stress test" our portfolio. If a 56% market drop happens again, I have some sense as to how our portfolio might behave (I emphasize might). There is no way to predict the future of course, nor any way for me to easily mimic the past using our portfolio of today.

Evidence, however limited, suggests our portfolio is likely to underperform in a fast growth environment. If a 130% (US) or 68% (CDN) market bounce occurs over a relatively short period of time, I would expect our portfolio to underperform considerably.

I am ever mindful of the trade off between "dead money" and performance drag vs. possible long-term catastrophic losses as we approach retirement age. I struggle with this quite a bit and can't say I have a fully satisfactory answer.

I have little doubt my portfolio allocations will evolve over time as I learn more, as we age, and as the real-world performance of our portfolio unfolds.

My next article will discuss how I both monitor holdings against targets and rebalance as needed. I will also touch on the kinds of considerations that might make me modify some or all of the target allocations over time.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: Our household portfolio consists of Canadian, US, international developed market and international developing market equities, Canadian bonds, bullion and cash.