The universal portfolio

Assertions:

1. Everyone has a portfolio.
2. Every portfolio has 5 categories.
3. Each category can grow and shrink independently of the other categories.
4. Each category is based on characteristics, not the thing itself.
5. Things may move between categories based on historical performance.

Let’s use dessert as a metaphor.

We’ll have a plate, there’s always a plate. We’ll have at least one brownie, there’s always one brownie. We may order another brownie. We can also add condiments such as whip cream, fruit compote, etc. Finally, we can add sprinkles like chocolate chips or similar.

The plate represents your liabilities and expenses, time or money you exchange with other people. Everyone gets a plate.

The first brownie is your short-term assets and income, the lifetime you have, or money you exchange with other people, usually for liabilities and expenses. Everyone has at least one brownie. For some people, the brownie might be small compared to the plate, their liabilities are greater than their short-term assets. For others, the brownie might be edge-to-edge and 4 times thicker than the plate. And all points in between.

The second brownie is what we call growth assets. Not everyone orders this brownie. For others, this brownie dwarfs the first brownie.

The condiments are negatively correlated assets. Compared to the second brownie (growth assets), things in this portion of the dessert tend to increase while growth assets decrease and vice versa. Not everyone has condiments. Other people are like that relative who takes a tiny sliver of pumpkin pie and adds two cups of whipped cream.

The sprinkles are assets with a low correlation to the other portions of the dessert. Again, some people really like sprinkles, possibly to the point of only having the first brownie and sprinkles. Other people avoid sprinkles entirely.

From a quantity perspective, you can have the same volume of stuff making up this dessert. However, when looking at the ratio, these components can shift. And they can shift for all sorts of reasons.

Let’s go deeper.

We’ll give the general characteristics of each, some things that fall into each category as of this writing, and how categories get larger or smaller in volume without necessarily changing the number of things in them.

Primary characteristics:

1. You give something you have to someone else once or regularly.
2. Historically, the real rate is the most negative compared to the other components.
3. Interest rates are in terms of APR and indicate excess money given to someone else.
4. Interest rates are working against your net worth.

By “historically,” we’re talking in terms of decades; 10, 20, 30, 40, and so on years. We typically give more weight to recent data as things change and evolve over time. What was true 40 years ago may not be true now, and including data from 40 years ago can hide more recent data due to averages and medians.

Real rate is adjusting the rate of the thing by inflation. You have an auto loan with a 4 percent APR. inflation is 2 percent. The real rate for the auto loan is 6 percent. Technically the real rate for the auto loan would be negative 6 percent because the APR and inflation are working against your net worth.

For assets, the interest rate is positive (working for you), but inflation is still negative. If you have a savings account earning 1 percent and inflation is 2 percent, the real rate of the savings account is negative 1 percent. Not as negative as the auto loan, which is why that savings account may not be classified as a liability in the portfolio.

Ingredients that go into making the plate:

1. Recurring bills: Utilities, rent, food, and similar purchases.
2. Terminal loans: Like a mortgage or car loan.
3. Revolving loans: Like credit cards and lines of credit.
4. Future replacement costs: A cellphone has a useful life of 2 to 5 years. A roof has a useful life of 15 to 20 years. After that time, you’ll probably want or need to buy a new one.

That’s not an exhaustive list. And the key characteristic is that you can’t eat a plate. The ingredients making up the plate go to someone else.

There are 4 major things that can make the plate grow and shrink; 2 are most likely out of your control:

1. Inflation and deflation.
2. An emergency.
3. The hedonic treadmill and lifestyle inflation.
4. The Diderot Effect.

Inflation is the tendency for the same quantity of products and services to cost more over time; we’re not discussing quality. This also goes back to the commentary on favoring recent data. Let’s look at the BLS Consumer Price Index data. We’ll use the “all items” category for each month.

From February 2022 through January 2023 (12 months), the average inflation rate was: 7.9 percent. From February 2013 through January 2023 (120 months or 10 years), the average was: 2.5 percent. From February 2003 through January 2023 (20 years), the average was: 2.49 percent.

There were some deflation months in 2009. 2009 was during The Great Recession, and inflation averaged 3 to 4 percent the previous year.

Notice that inflation for the last year is almost triple the last 10 and 20 years. That’s because the data from 10 and 20 years ago is dragging down (or hiding) the most recent data. And I don’t know about you, but the roughly 8 percent impacts present-me much more than the percentages from 10 and 20 years ago. Again, a reason for giving more weight to the recent data.

Regardless, the plate has increased by almost 8 percent due to inflation, and it’s out of my direct control and influence.

In 2022, I had appendicitis leading to an appendectomy. My plate got bigger because of an emergency.

Hedonic adaptation is more within your control. You get a raise. You buy a more expensive thing. You’ve made your plate bigger because you have to replace that thing, and if you replace it with the same thing, the price may be the same, but it’s more than the first thing you replaced (this is why we can’t have nice things).

The Diderot effect is when you buy a new pair of shoes. Now you decide to replace your wardrobe because all your clothes are old compared to your shoes. Then you decide to replace your furniture because your new clothes seem so out of place compared to the sofa you bought 15 years ago with multiple stains. You’ve made the plate bigger.

Primary characteristics:

1. Performs better in deflationary and contracting environments.
2. Someone gives you something they have once or regularly.
3. Historically, the real rate is negative or neutral compared to the other categories.
4. Interest rates are in terms of APY and indicate excess money given to you.

A non-exhaustive list of ingredients that go into making the first brownie:

1. Cash (includes your paycheck): The ubiquitous method of exchange for the economy you’re participating in. Could be the USD, precious metal coins back in the day, or cigarettes in those old prison movies.
2. Short-term bonds: Bonds with a 1 to 3 year maturity date.
3. Certificates of deposit and share certificates: Cash held at a bank for a period, usually less than 10 years.

Some things that cause this brownie to get bigger and smaller:

1. Saving or reinvesting dividends.
2. Inflation and deflation.
3. An emergency.
4. The hedonic treadmill and lifestyle inflation.
5. The Diderot Effect.

Yep. Almost the same list as the plate.

This reveals the notion of spending less, earning more, or both.

Some might say compounding has the greatest impact, but this isn’t the space for that complexity and debate. It’ll be covered in a later section. Further, compounding falls under the “making more” part. But let’s start with spending less as we tend to have more control over that.

A recommended precursor to spending less (and most things) is to get an idea of where you are, also known as tracking. Capture how much short-term assets you spend (the brownie you eat) and why you spent it. Focus on the categories you spend the most on and see if you can reduce or eliminate them.

The plate gets smaller, and the brownie may increase in volume.

The truth about making more is that not every employer or client will increase your income annually. There aren’t mechanisms that cause regular increases. Sometimes, you must advocate for yourself or change how you earn income; otherwise, you are taking a pay cut because some aspects of the plate don’t care about your situation. The volume of the brownie remains the same, but the plate gets bigger.

This is one reason I include liabilities in the portfolio. Paying off liabilities takes a bite of the brownie while making the plate proportionally smaller. However, it doesn’t actually change the proportion of the portfolio.

I have 5,000 USD in liabilities. I have 5,000 USD in short-term assets. My ratio is 50:50. I spend 4,000 USD from short-term assets to pay a liability. My liabilities are 1,000 USD, and my short-term assets are 1,000 USD; still 50:50.

To change the allocation, we must change something else; spend less, earn more or both.

Go ahead and pay off the credit card, but don’t run it up again; spend less. Be good at what you do and ask for a raise or increase your prices; earn more. Or both.

Let’s say I pay off the 5,000 USD using the 5,000 USD. The plate is still there (mainly due to inflation and other bills), and I have no brownie. Then I earn 1,000 USD. My first brownie is back. Let’s say I have 200 USD in bills; the portfolio is about 20:80. I pay the bill, and the portfolio is 0:100.

Some people are in a position where the phrase “feast or famine” truly applies. Further, in some cases, they can’t see alternatives. Finally, sometimes they can see an alternative but feel it doesn’t apply to them or they will fail (including spending less or earning more).

The last bit is about the environment. Bridgewater and Associates identified two spectrums for economic environments:

1. growth to contraction and
2. inflation to deflation.

This gives us the baseline for a quad chart of 4 economic environments, I’ll use a list instead of a chart:

1. Inflation and growth.
2. Deflation and contraction.
3. Inflation and contraction.
4. Deflation and growth.

Each of these environments has a probability of happening. And each type of asset tends to do better in one of these environments. Diversification, in this model, comes from choosing uncorrelated assets that perform well in each type of environment, favoring those with a high probability of occurring and the one you’re currently in.

Next brownie!

1. Performs better in growth environments. (This includes resell value, the ability to purchase cash and other short-term assets.)
2. Historically, the real rate is positive and higher than the other categories.
3. Interest rates are in terms of APY and indicate excess money given to you.

Secondary characteristics:

1. You typically convert short-term assets into something else you then own, not a liability.
2. Has a resell value that fluctuates more than short-term assets; a dollar is a dollar, and it’s rare someone is willing to give you 2 dollars for 1.
3. When you sell the thing, it becomes a short-term asset. Even if you sell with the intention of immediately purchasing another asset.

A non-exhaustive list of ingredients that go into making the second brownie:

1. Corporate equities: Ownership shares in companies publicly and privately held.
2. Some forms of tangible things (real estate, artwork, and antique cars, for example).
3. Mutual and index funds in various forms that purchase corporate equities: You purchase a share in the fund, and the fund’s custodian purchases shares in companies based on some guidelines and guardrails established for the fund.

Some primary things causing this layer to grow and shrink:

1. The risk premium: Where we get the idea of greater risk, greater return, which isn’t a hard truth; correlation is not causation. Just because it’s risky doesn’t mean there will be a reward (just look at most things we consider gambling).
2. Selling some or buying more of the things.
3. Appreciation and depreciation of those things.
4. Reinvesting dividends.

We’re buying something other than short-term assets, one reason this is a different brownie. Short-term and growth assets tend to have a low or negative correlation with one another, the definition of diversification.

That said, it’s important to note and remember that things can shift. Something you thought was a growth asset can turn into a liability or short-term asset based on the changing characteristics of the thing.

You can charge whatever you want for the thing when you sell it. Some questions this raises:

1. When do you need the short-term asset from the sale?
2. Will someone pay what you’re asking?
3. If so, will they do it in a timely manner?
4. Will you need to replace the thing with something similar?

This brings up the idea of liquidity. Liquidity is the time and cost it takes for something to be exchanged for something else. Short-term assets are considered very liquid, mainly because cash is an ingredient in that brownie. Growth assets usually have low liquidity. Low liquidity is a risk because you may be unable to “get at the money” when you need it. The risk premium at work.

Back in the day, and I mean the 1970s and earlier back in the day, buying shares in a publicly traded company was pretty prohibitive because of the lack of liquidity.

We’re talking old-school here.

Let’s assume you could buy direct. You would give someone your money. They would generate a piece of paper representing your share in the company. You’d wait to get the paper. You would then have to hold it as proof of ownership. If the company ever decided to pay its shareholders, it must keep records of your ownership stake and cut you a check. You would also need to keep tabs in case you missed getting your cut, then use the paper to verify your claim of ownership. (Shares are like the corporate equivalent of printing money or a community currency).

If you wanted to sell the share, you’d have to find someone willing to buy it. They’d have to give you money. You’d have to transfer ownership to this other person. Then give them the paper and let the company know to update their records. (Like transferring the title for a car or house.)

If you had to go through a broker (most times you did), you might have to pay fees of hundreds of dollars to pay them to do all of that for you.

In the twenty-first century, there are more options with varying fees ranging from 0 to a couple percent. Publicly traded company shares have become more liquid. And the old-school way still exists; old doesn’t mean bad or extinct.

Compare that to a tangible asset like a painting. Again, you purchase the thing. You wait for the thing. Sometimes there’s a letter of authenticity you’ll want to keep. Now you want to sell the painting. How long does finding someone willing to pay what you’re charging take? Again, risk.

Compare both to something you own and use, like a house. Historically, houses are growth assets. But, when you sell the manmade cave, you usually want to find another one.

Regardless, the second brownie gets a little smaller in volume once sold.

Let’s go back to risk for a second.

We often conflate risk with volatility. Volatility is how much the thing you bought appreciates or depreciates in a given period. Volatility is great when it’s going up and sucks when it’s going down. The risky side of volatility is needing to sell and having to charge less than your cost basis for the thing being sold.

Cost basis is what you paid for the thing plus any additional related costs for maintenance and improvement; also known as the total cost of ownership. Just because you paid a lot for something doesn’t mean someone later will pay your initial cost basis, much less anything extra.

I buy a car for 10,000 USD. The chances of me selling it for 10,000 USD later is pretty slim. The chances of me selling it for 10,000 USD, plus the taxes I paid over the years, the interest paid on the loan I had at the time, the oil changes, tires, and so on, are even slimmer. When it comes to cars, we don’t look at them like an asset, we look at them like a liability. We tend to look at homes as assets because we don’t keep track of all the money spent over the course of ownership. Of course, we tend not to track total cost of ownership for most things.

Many growth assets will pay some form of income, whether dividends or rent. The dividend and rent you receive, strictly speaking, is a short-term asset.

Primary characteristics:

1. Performs better in inflationary and contraction environments.
2. Historically, are more negatively correlated to growth assets.
3. Interest rates are in terms of APY and indicate excess money given to you.
4. Historically, the real rate is lower than growth assets.

Secondary characteristics:

1. Same as growth assets.

In the United States, the asset type with the highest negative correlation historically to corporate equities is:

1. Long-term US treasuries: A loan to a government with a maturity date of around 20 or more years.

Primary things causing this layer to grow and shrink:

1. The risk premium.
2. Selling some or buying more of the things.
3. Appreciation and depreciation of those things.
4. Reinvesting dividends.

Yep. Same as the list for growth assets.

Many negatively correlated assets have a lower risk premium. The things that fall into this part of our dessert also tend to be less volatile.

When folks talk about a stock-bond portfolio, the benefit isn’t in the type of asset, specifically company stocks and any old bonds.

Humans have a tendency to try and expedite communication by cutting out the parts of speech we think are superfluous. So, instead of saying, “A diversified collection of stocks in relatively stable companies and long-term treasuries,” we just say, “stocks and bonds” and hope the other person unpacks it as we intended.

There are all types of companies, and they all have at least one share. They’re not all growth assets; most will fail (“go to zero”).

There are all types of bonds. Corporate bonds, state municipal bonds, and federal government bonds. Short-, long-, and intermediate-term. Then there’s the credit rating for the bond issuer; triple A, double A, and so on.

We say stocks and bonds as shorthand for growth and negative correlation. We want something with the characteristics described. Sticking with an asset type, despite a change in historical characteristics, would be a foolish consistency.

Corporate equities historically have a positive real rate of return. Historically, negative correlation assets increase their real return rate when growth assets are decreasing in theirs. If this relationship changes over a long enough period, we might want to look for a different asset type. We’ll go into depth on correlations in a later section.

Primary characteristics:

1. Performs better in inflationary environments.
2. Historically, they have a low correlation with things in the other categories.
3. Interest rates are in terms of APY and indicate excess money given to you.

Secondary characteristics:

1. Same as growth assets.

In the United States, two asset types that historically fit these characteristics are:

1. Commodities: A basic good often exchanged for similar goods. The goods include metals, energy, livestock and meat, and agriculture.
2. Gold: A specific type of commodity that has been historically used as currency (or to back currencies); no longer the case in most contemporary economies. Gold does have some utility in electronics.

Primary things causing this layer to grow and shrink:

1. The risk premium.
2. Selling some or buying more of the things.
3. Appreciation and depreciation of those things.

Almost the same list as growth and negative correlation, but notice we took out reinvesting dividends. Commodities typically don’t pay a dividend. Further, when you sell them, they may be taxed differently than other types of assets. In this sense, they are less investment and more speculation according to the definitions laid out by J. David Stein.

Back to the compounding debate and how much of a contributor it is.

If I give you a dollar every day for 50 days, that’s your paycheck. That’s linear (we’ll start skipping after the first 10 days, so keep track of the day column):

That’s not bad. Double your money from day 1 to day 2. However, on day 3, the increase is only 50 percent. On day 4, the increase is 30 percent. On day 5, it’s 25 percent. On day 6, it’s 20 percent. Day 7, it’s 16. When you have no money and are thinking 1 day at a time, every bit feels like a million.

Humans tend to have a difficult time thinking long-term and exponentially. We’re much better at short-term and linear.

If I give you a dollar and only 10 percent of the amount you keep every day thereafter, that’s exponential:

From day 1 to day 2, it’s kinda, “meh.” Even up to day 6, you might think, “I’d rather have the dollar a day” (indicative of a steady paycheck and short-term linear thinking). Heck, it’s not actually interesting until days 25 to 30 because that’s the tipping point.

That’s where the 10 percent is more than you earned each day in the linear example. Then 5 days later, you’re increasing by over double what you earned in the linear example. 5 days later, it’s triple what you were increasing in the linear example, and you have roughly the same balance; you have caught up to your earnings from the linear paycheck example. 10 days later, it’s 9 times the increase of the linear example and twice the balance.

The other thing we can take away from the exponential table is what’s referred to as the rule of 72. Basically, divide 72 by any rate of return (APY), and you will arrive at roughly the number of compounding periods it will take for a specific dollar amount to double. In our case, 72 divided by 10 is 7.2. It took about 8 days to go from 1 dollar to 2. It took about 16 days to go from 2 dollars to 4. We hit 8 dollars around day 20. And so on.

So, yes, compounding is very powerful. It also takes two things to really kick in:

1. time and
2. money to put somewhere to compound.

1 dollar at 1 percent APY in a regular savings account will double in 72 years. If I live hand-to-mouth, there aren’t even seeds left to plant at that 1 percent return.

My regular dentist recommended someone else when it came time to remove my wisdom teeth. When I didn’t react well to the estimated cost the other dentist quoted me, they said, “There are always places you can cut in your budget.” Sure, I could have bitched about healthcare in the United States and my lot in life working slightly over minimum wage with health and dental insurance and how “the system” is out to screw us. But I didn’t.

Instead, I laughed the laugh to keep from crying. I went to my car and did cry. I returned to my regular dentist, told him the story, and shed a tear or two in the process. He said, “That was pretty elitist. I guess I won’t be recommending them anymore. The two teeth on the bottom are pretty far out, and I can probably pull them. We’ll bill it as an extraction. The ones on the top are pretty impacted and probably not going anywhere, and I don’t want to get near your sinuses. We’ll leave those where they are and check on them regularly to ensure they haven’t moved.”

We extracted the teeth. Almost 20 years and 3 more dentists later, the top wisdom teeth haven’t moved. And none of the dentists have recommended even trying.

It’s difficult for humans to think long-term and exponentially. It’s also difficult to understand and empathize with situations that aren’t ours or those closest to us.

Two of the asset categories are in terms of correlation. Correlation is measured from -1 to 1. -1 means one of the things moves in the opposite direction of the other. 1 means both things move in the same direction. 0 means we’re unsure in which direction one of them will move.

Let’s look at correlations of one asset type in each part of the dessert, not including the plate. We’ll use the following proxies inside an asset correlation tool provided by Portfolio Visualizer:

1. First brownie (short-term assets): Cash.
2. Second brownie (growth assets): A popular total US Stock Market index fund (VTSAX).
3. Condiments (negative correlation assets): Long-term US Treasuries (TLT).
4. Sprinkles (low correlation assets): Gold (GLD).

We’ll only be able to do 10 years and roughly 19 years because the gold fund we’re using started in 2004.

The hypotheses going in:

1. Cash will have a 0 or slightly negative correlation with our representative growth asset.
2. Growth assets will have a 0 or negative correlation with the other three.
3. Long-term US Treasuries will have the greatest negative correlation with the growth asset.
4. Gold will have a close to 0 correlation with the other three.

For 10 years, we’ll go from February 2013 to the end of January 2023, here’s what we find:

1. Cash is negatively correlated with the growth asset at -0.07; hypothesis confirmed.
2. Growth assets range in correlation from -0.07 (compared to cash) to 0.06 (compared to gold); hypothesis confirmed.
3. Long-term US Treasuries are negatively correlated with the growth asset at -0.03; this does not confirm the hypothesis.
4. Gold ranges in correlation with the other three from 0.06 (compared to the growth asset) to 0.4 (compared to long-term US Treasuries); this does not confirm the hypothesis.

In this case, there are no assets meeting the negative correlation characteristics; more on that later.

Does this mean we should shift our thinking? Is this the proverbial paradigm shift you may have heard about? Do we need to find something else to replace long-term US Treasuries as a negative correlation asset before adding condiments to our portfolio? Or, if we already have long-term US Treasuries, should we rebalance into something else with a greater negative correlation?

Maybe. Only time will tell. Let’s look at a period longer than 10 years and see what we find.

For this, we’ll use the period from February 2005 to the end of January 2023, and using the same hypotheses, here’s what we find:

1. Cash is negatively correlated with the growth asset at -0.06; hypothesis confirmed.
2. Growth assets range in correlation from -0.19 (compared to long-term US Treasuries) to 0.08 (compared to gold); hypothesis confirmed.
3. Long-term US Treasuries are negatively correlated with the growth asset at -0.19, greater than the negative correlation of cash; hypothesis confirmed.
4. Gold ranges in correlation with the other three from 0.08 (compared to the growth asset) to 0.23 (compared to long-term US Treasuries); the hypothesis is confirmed.

In this context, if something has a correlation of 0.4 or greater (40 percent or more), I wouldn’t call it a low correlation, and it’s definitely not negative. If something has a negative correlation between 0 and -0.1, I wouldn’t say it was negative enough to be in the negative correlation category.

For the last 10 years, it seems like maybe we are seeing a shift in asset behavior. But, for almost 20 years, we’re seeing the hypothesized behavior. The new behavior might indicate a shift, and we should keep an eye out. But until there are more periods showing where the original hypothesis failed, I wouldn’t make any drastic changes to anything.

As of this writing, I’m still in accumulation mode and only have the plate and first two brownies occupying any significant portion of my portfolio. And I still have a few years before seriously contemplating adding condiments to my dessert.