In the last post about investing, we showed that investing is a powerful way to address the long-term risk of inflation eating away the value of our savings. We also discussed the difference between stocks, bonds and other types of investments, and drew some distinctions between investing and speculation.

• Investing: taking broad risks inherent to capitalism (competition for customers, the business cycle, etc.) that we know empirically (the data backs it up) have compensated investors for hundreds of years.
• Speculation: taking narrow risk that one specific company or unproven outcome will prevail. Often a new technology, regulatory outcome, or other change from the status-quo needs to happen for it to be profitable.

If I had to pick the one most powerful thing someone can learn to make them a better investor, it would be that investing and speculation are different things. Speculation gets more media attention than investing does, but it’s often harmful to wealth building, whereas investing done right is beneficial.

Benjamin Graham, the father of value investing and Warren Buffet’s mentor, owns the definitive quote on the subject:

“An investment operation is one which, upon thorough analysis, promises safety of principal and an adequate return. Operations not meeting these requirements are speculative.”

Benjamin Graham, The Intelligent Investor

Graham’s quote offers another powerful distinction between investing and speculation:

• Investing involves a process for analyzing, buying, and selling investments.   
• Speculation is devoid of such process, and relies instead on stories and emotion (“Bitcoin will become the new monetary standard because X, Y, Z”)

Ben Graham had a meticulous process for evaluating stocks (which he taught to Mr. Buffett), and a high level of confidence in his abilities. So we can forgive him for using the word “safety”, which has no place in a conversation about investing. All investing involves risk. Investments are not “safe”. However, due to the long-term risk posed by inflation (link), cash under the mattress may not be “safe” either. Following a process for investment decision-making can help manage (but never eliminate) risk.

What might such a process look like? That’s what this post will address.

I’m going to take you through a process that institutions like large advisory firms, pension funds, and endowments use to build portfolios. It’s a process I’ve used in my career and found to be a powerful way to develop investment proficiency.

*** This is for educational purposes only and is not investment advice. I don’t know you or your personal situation, so I can’t say what investments may or may not be appropriate for you. Please do your own research before investing in anything you read about online. ***

With these important caveats in mind, let’s jump in.

🔖👇👇

what’s the goal?

Why do we invest? To earn a good return on our money so we can build wealth. Wealth building happens when investments have time to compound, or earn interest on top of interest. The better the returns of the investments in our portfolio, the more powerful they will compound and the more wealth we build. So we want attractive returns because that helps us realize our investment goals (retirement, Lamborghini, etc).

The catch is: attractive returns come from taking risk. If we avoid risk, we should likewise expect lower returns. Look at the difference in the 100-year returns for stocks vs. T-bills in the below chart from a prior post for example.

As we venture out on the risk spectrum seeking higher returns, we steadily add more and more risk to our portfolio. More risk means the value of our investments fluctuate more, which can make it hard to stay invested. Look at how the graph for stocks bounces around compared to that for T-Bills. It can be difficult to watch the value of our nest egg bounce up and down, causing stress and unpleasant emotions. So we also want to limit the amount of risk we take so we can sleep at night.

So the two broad goals we start with are attractive returns and managing the risk we take to achieve them. The difficulty comes in balancing these two conflicting objectives. At the end of the day a good investment policy should provide clarity around the return we can expect our portfolio to deliver, and the risk we expect to take in getting there.

balancing riks & return objectives

We can address this problem in two ways:

• Estimate how much risk we can handle and calculate the historical market return associated with that risk level (start with risk, end with return).

• Estimate the return we need to hit our goals, and build a portfolio likely to deliver it (start with return, end with risk).

To maximize the chance of hitting our financial goals, we should do both. Estimating the level of risk you’re comfortable with is often done by talking with an experienced advisor or filling out questionnaires. This is a crucial step, which we will address in future posts.

The second method involves two pieces. Estimating the return needed to reach our goals is simple:

• How much do the goals cost?
• When do you need the money (Years? Decades?)
• How much money do you have today?
• What growth rate allows current funds to grow equal to future costs over the designated timeframe?

We can even write it as an equation if we want to get super nerdy:

Cost of goals = (current funds)*(1+rate of return) ^ # of periods

Goals are personal for all of us. Like I said above, I don’t know you so I won’t make assumptions about your goals or what it will take to fund them. This is literally what a good financial advisor does and if you have one, they are (hopefully) helping you clarify your goals and calculate what they are likely to cost.

“expected” investment returns

Let’s focus on the “rate of return” piece for now. Unlike goals, which are unique to every investor (like your eye color), the returns offered by capital markets are experienced by all investors. The nature of markets change over time and so do the returns investors can expect from different types of investments (stocks, bonds, “other”). Think of the level of returns investors expected and experienced during these time periods:

• The 1990s: economic growth, rising stock market, falling interest rates, falling gas prices, rise of the internet.
• The “lost decade” 2001-2011: “dot-com” bubble, 9-11, WorldCom and Enron accounting scandals, real estate bubble, 2008 financial crisis
• 2011 – 2020: low interest rates, “easy money” policies, EU sovereign debt crisis, de-globalization (Brexit, Trump tariffs), rise of cloud computing and cryptocurrencies
• 2021 – 2024: global pandemic, economic shut-down and re-opening, large government stimulus, inflation, rise of artificial intelligence

With the benefit of hindsight, we know what returns stock and bond markets provided in each of these periods. Depending on which areas of the market someone invested in – which asset classes – they could have done better or worse than the overall market. But a sufficiently diversified portfolio, held over a full market cycle (encompassing both “bull” and “bear” market environments) can expect to achieve something like the market rate of return.

Professional investment firms seek to define in advance the level of returns and risk they expect from different types of investments given the current economic environment. They use those estimates to build portfolios via a process called “strategic asset allocation” (SAA for short). The primary goal of SAA is to build a portfolio that can survive multiple types of market environments.

Let’s go through the process to see how it works.

what is asset allocation and why should you care?

Before we dive in – why should you care about this?

• DIY investors: SAA can help you build a portfolio that balances your risk AND return objectives. It can help avoid falling into speculation.

• If you pay an advisor to invest your money and they aren’t doing this (or something similar), this can help you decide if it makes sense to keep paying them.
• If you’re an advisor, this is a great way to document your investment process used for clients. Handy to have on file for clients and regulators.

SAA involves three steps:

• Define asset classes
• Gather market data
• Create portfolios

step 1 – define asset classes

“Asset classes” are a nerdy way of referring to categories or types of investments. Some examples:

• Stocks
• Bonds
• Real estate

We can get more specific and define “sub-asset classes” like:

• US stocks
• international stocks
• investment-grade bonds
• high yield “junk” bonds

Like the way one might separate kitchen utensils or our children’s toys into different categories (forks, spoons, knives – blocks, dolls, crafts), defining asset classes simply divides the group of investment opportunities available to us into categories with similar traits.

PORTFOLIOS BUILT TO LAST

Defining asset classes is effectively saying “these are the investments my portfolio will hold”. We want to make sure the portfolio we end up with:

• is exposed to multiple types of risk
• avoids/minimizes speculative investments
• reflects most investment opportunities available at any point in time
• can withstand different market environments that may come to pass

Think about the global investment market, as it stands today and how it may change in the years ahead. The ideal portfolio contains assets that can benefit from different market backdrops which may come to pass.

• Strong economy (2021 COVID recovery, 1990s)
• Recession (2008, 2001)
• High inflation (the 1970s, 2021 – 2022)
• Low inflation (the 2010s)
• Rising interest rates (2018, 2022)
• Falling interest rates (1980-2015)

To identify such investments, let’s review some different asset classes and the risks to which they are sensitive. This will allow us to gauge which ones might be valuable to include in our hypothetical portfolio.

USING RISK TO SELECT ASSETS

Let’s go back to our list of broad asset classes.

• stocks
• bonds
• real estate

What risks are they sensitive to?

STOCKS

Stocks are ownership interests in companies. Investors get a share of earnings (if there are any). Stocks do well when companies make money. Companies make money when the economy is growing and people have money to spend. Economic growth and moderate inflation tend to be good for stocks (they can pass on price increases to customers). But prolonged high inflation can be bad for stocks (higher material and labor costs eat away profits).

Easy credit conditions and low interest rates allow companies to borrow freely and finance their operations. This helps profitability. Also, some investors use borrowed money to purchase stocks. So lower rates and easy credit conditions help buoy the stock market.

BONDS

Bonds are contractual agreements. Issuers must make payments regardless of what the economy does. Stockholders get whatever is left after bondholders get paid. Companies may have to stop paying stock dividends and prioritize interest payments in a downturn. For this reason, high quality bonds can help during economic downturns.

Because bond payments are fixed, inflation is a big risk. Payments made later in the bond’s life are worth less than those made earlier.

Bonds are sensitive to interest rates. Lower rates make previously issued bonds (paying higher interest) attractive to investors. As rates rise, bonds issued at prior (low) rates become less attractive. This is the price component.

But consider the income component; bonds issued at higher market rates will pay more income over their life vs. bonds issued in low-rate environments. Thus, the longer rates remain high, the income benefit may outweigh the price impact.

REAL ESTATE

Real estate, like stocks, tends to do well when the economy grows. People can afford to rent or buy homes, and companies may need to expand their operations. Real estate can be hard to sell during a downturn.

Real estate investing often involves the use of debt (AKA “leverage” if you want to sound smart). Thus, like bonds, real estate tends to be sensitive to interest rate moves. High rates make it tough to finance deals, so fewer transactions occur. Low rates tend to spur the real estate market. This is one reason policymakers took interest rates to zero after the housing market collapse and resulting recession in 2008.

Real estate can do well if inflation rises because owners can raise rents. Real estate transactions tend to be large and stressful (packing up and moving your home or business). They’re an ordeal, so people will often put up with higher rent, HOA fees, etc. rather than move.

We could continue this exercise much further – there are tons of risk exposures in markets, but to keep this example simple, let’s stop here and summarize our asset classes and their risk sensitivities:

STOCKS

Economic growth (very good), inflation (moderate is good; high is bad), credit risk (matters), interest rates (lower is good)

BONDS

Economic growth (less important), inflation (bad), interest rates (lower is good for prices; higher is good for income)

REAL ESTATE

Economic growth (good), inflation (less important), credit availability (important), interest rates (lower is good)

Stocks give us exposure to economic growth, a powerful force in our capitalist system. Bonds help buffer the risk of negative credit and growth environments. They also offer income if rates remain high. Real estate somewhat addresses the risk of inflation, which can be bad for both stocks and bonds. Again, this is a simplified example, and obviously one could bring up other consideration for our simple set of asset classes. But given what we’ve outlined here, these could be complementary if held together in a portfolio.

We’ve defined our asset classes (high five!) On to the next step.

STEP 2 – GATHER DATA

Now that we have our asset classes, we need data so we can see what kind of returns they may be able to generate for us.

We can look at historical data, and this is helpful, but we are investing in the future, not the past. It would be ideal to know what kind of growth we might expect from our assets GOING FORWARD.

Unfortunately, as Yogi Berra famously said, “making predictions is hard, especially about the future”. We don’t know for sure how our assets will perform going forward. Rats.

Here’s a thought – what if we had a team of PhDs, economists, and analysts who did nothing but study investments, and we put the problem to THEM? Get excited because you absolutely DO have such people’s expertise at your disposal.

ask the experts

Many investment companies publish estimates for what different asset class returns will be in the future. Because it’s finance, there must be a fancy term for such estimates and that term is, “Capital Market Assumptions”. Do a Google search for CMAs and you’ll find a bunch.

One such particularly useful publication is the Horizon Actuarial Survey (you can tell from the name it’s exciting!) At the risk of putting you to sleep, Horizon is a consulting firm that works with pension plans. Here’s a link to their data, which they update every year:

https://www.horizonactuarial.com/survey-of-capital-market-assumptions

The data we want is summarized toward the end of the latest release:

WHERE DO THE ESTIMATES COME FROM?

The list of the 42 investment firms Horizon surveyed to compile their CMAs is on page 3 of the release. You may recognize some of the names on this list, which includes Vanguard, JP Morgan, and Goldman Sachs.

From each of these companies, Horizon gathered FORWARD-LOOKING estimates for return, risk and correlations (degree of co-movement) of various investment asset classes. These numbers are not guarantees – no one can know the future – but rather the combined views of 42 teams of professional economists and research analysts who study markets for a living. These companies have large teams, powerful software, and lots of experience. You can and should look up the individual firm CMAs, which, in many cases, are published for free online. Here’s a link to JP Morgan’s:

https://am.jpmorgan.com/us/en/asset-management/institutional/insights/portfolio-insights/ltcma

I don’t know about you, but I can’t call up Dr. David Kelly (JP Morgan’s Chief Global Strategist) to ask his thoughts on the stock market. But those thoughts are absolutely reflected in JPM’s capital markets assumptions, which in-turn are part of the Horizon survey.

Horizon lists the EXPECTED average (arithmetic) annual returns for our assets over the next 10 years in the first column:

• Stocks of large US companies (US Equity Large Cap): 8.2%
• High quality bonds (US Corporate Core): 4.9%
• Real estate: 7.3%

Will the actual returns we get equal these “estimated” returns? It’s unlikely. Not even an army of PhDs can reliably predict the future (remember Yogi Berra – it’s hard). But these thoughtfully-constructed estimates incorporate the views of some of the most experienced and intelligent professional investors in the world.

We want to get as many different viewpoints as possible into our estimates. Doing this makes it more likely that our portfolio will stand the test of time. This is a great aspect about the Horizon survey; it combines estimates from multiple firms, so no one viewpoint is emphasized.

We could do worse than use such views as a starting point when building portfolios. It’s better than sticking a finger in the air.

Next we’ll explore the data a bit more, and start to play with what a portfolio might look like.

QUICK RECAP

So far we’ve learned:

• Investing is all about balancing risk and return.
• One way to do this is by estimating the return we need to hit our goals, then building a portfolio that analysis SUGGESTS MIGHT deliver that return (no guarantee).
• Strategic Asset Allocation (SAA) helps us identify what types of investments we want in our portfolio (asset classes) and analyze data from experts to build portfolios targeting specific levels of EXPECTED (not guaranteed) returns.

Now we need to crunch the numbers and see what returns a hypothetical mix of the three-asset classes above might offer, as well as how risky such a portfolio might be. If these risk-return estimates meet our personal goals, we could implement them and have a chance at success. Remember, this is for educational purposes and not intended as advice or a recommendation.

Let’s continue reviewing Horizon’s 2023 Capital Market Assumptions.

behind the NUMBERs: RETURNS

Horizon publishes two types of expected returns: arithmetic and geometric. Arithmetic returns are simple averages. Geometric returns are growth rates that equate starting and ending values.

For this exercise, we’re interested in building long-term investment portfolios to reach our goals. We’ll use geometric returns (second column).

We’ll use the 10-year return assumptions (fist set of returns) because it’s a realistic time period over which we might hold our hypothetical portfolio.

behind the NUMBERs: RISK

Investment professionals use the statistical term “standard deviation” to measure risk. Standard deviation refers to how far a specific observation is from the average. In investing, we can think of it as how far this year’s (month, quarter, etc.) return is from the long-term average. Investments with large variations tend to be more risky – up one year, down the next, etc.

Smaller variations imply more stable investments over time. The odds you can meet future spending needs are higher. So a lower standard deviation = less EXPECTED risk (remember – these are uncertain predictions. The future will be different).

Finally, we have correlations listed in the triangle-shaped table to the right. In investing, correlation refers to the degree one investment moves when another moves.

The highest value is 1 (perfect co-movement in the same direction). 📈📈

Zero correlation means the two assets have no statistical relationship. 📈🤷‍♂️

The lowest is -1 (perfect co-movement in opposite directions). 📈📉

GATHERING ASSET CLASS DATA

Let’s assemble the risk-return data for our asset classes from page 14 of the Horizon survey:

Stocks (large US companies)

Expected 10-year geometric return: 6.9%

Expected risk (standard deviation): 16.64%

Correlation vs. bonds: 0.26

Correlation vs. real estate: 0.56

Bonds (US corporate core):

Expected 10-year geometric return: 4.71%

expected risk (standard deviation): 5.85%

Correlation vs. US large stocks: 0.26

Correlation vs. real estate: 0.25

Real estate:

Expected 10-year geometric return: 5.95%

Expected risk (standard deviation): 16.72%

Correlation vs. bonds: 0.25

Correlation vs. US large stocks: 0.56

Great, we’ve finished step two of SAA, gathering data (high five!) Let’s build some portfolios.

BUILDING PORTFOLIOS

To build portfolios we’ll need to assign percentages to our three assets (stocks, bonds, RE), and then calculate the expected return and risk for the resulting mix.

Let’s start by finding portfolios with the highest expected return and lowest expected risk, and then we’ll fill in the middle.

HIGH EXPECTED RETURN PORTFOLIO

Remember, in SAA the asset classes we defined in step one represent the entire investment opportunity set. In this example, the portfolio with the highest EXPECTED return is 100% stocks because, of our three assets, stocks have the highest expected return. An all stock portfolio has an expected return of 6.9% and expected risk of 16.6% (no calculations needed, it’s just the risk-return estimates for asset class). As we start to mix bonds and real estate in with our stock portfolio, our expected return goes down.

PORTFOLIO MATH: EXPECTED RETURN AND RISK

Once we have more than one asset class in our portfolio, we have to do some math (sorry). The expected return for a portfolio of assets is a simple weighted average:

[(% allocation to asset 1) * (asset 1 expected return)] + [(% allocation to asset 2) * (asset 2 expected return)]… repeat for all portfolio assets.

For example, the expected return for a 50/50 stock/bond portfolio would be:

(0.5 x 0.069) + (0.5 x 0.0471) = 0.058 or 5.8%

Portfolio risk is more difficult to calculate. We’ll enlist the power of Excel to help.

Portfolio standard deviation = Square root [ (allocation to asset 1 squared) (std dev of asset 1 squared) + (allocation of asset 2 squared) (std dev of asset 2 squared) + 2 (allocation 1)(allocation 2)(correlation between asset 1 & 2)(std dev 1)(std dev 2)]

This link explains the math in more detail.

Using the above formula, expected risk for a 50/50 stock/bond portfolio is 8.82%.

Sidenote: check out CFA Institute if you are enjoying this!

LESS RISKY PORTFOLIO

Ok, now what’s the least risky portfolio we can build with our three assets? We can use Excel’s Solver add-in to help.

Here’s how to load the Solver add-in into Excel

Open Solver by clicking the “data” menu at the top of the screen in Excel (you can also open it from the “home” menu, it looks different but will calculate the same), and enter in the objective and constraints.

• Objective (what we’re asking Solver to do): minimize portfolio risk (cell B13)
• Variables (values we’re asking Solver to change until it hits the goal): allocations to stocks, bonds, RE (cells H3, H4, and H5)
• Constraints (Solver cannot violate these in seeking the goal):
  • total portfolio allocation (H7) must add up to 100%
  • asset class allocations must be positive (H3:H5 >= 0)

The LEAST risky option is not to invest at all. But that also leads to zero return. So we’ll require Solver gives us allocations that add up to 100%.

Solving for this scenario, we get a portfolio allocation of 10% Stocks, 80% Bonds, and 10% Real Estate. The portfolio’s expected return (B9) is 5.0% and its expected risk (B13) is 5.24%.

Compare the stats for our least risky portfolio with holding 100% bonds, the least risky asset class. Bonds expected return = 4.71%, expected risk = 5.85%.

This is where diversification shows its value. By mixing in stocks and real estate with bonds, we actually DECREASE the expected risk of the portfolio. How can this be, when both stocks and real estate are more risky individually?

The answer is in the correlation, or co-movement; despite the higher variation (more risk) of stocks and RE, they do not move in lockstep with bonds. Actually the correlations of 0.26 and 0.25 are quite low (closer to zero than to 1). We expect the risks of these three assets to express themselves at different times, and to different degrees.

Stocks might behave “risky” at a time when bonds behave well, and vice versa. Assets with imperfect co-movement (correlations below 1.0) help to offset each other’s risk. This is a foundational concept in investing, and the asset allocation process helps us see mathematically why it makes sense.

Ok let’s pause here. In the next post we’ll build more portfolios using Excel’s Solver function.

Thanks for reading, I hope you found it valuable. Join the conversation on Twitter and subscribe to updates using the buttons below.

Wishing you wealth and peace.

– Matt

🙏💵🙌 empowermenttools.net