How a Portfolio Is Built

Start with what the money has to do

The purpose of a portfolio decides its shape before anything else does. Consider two people. A couple who sold their business last year wants to keep what they have, live on it, and hand it down in good order. A surgeon with the same balance is still building, because his number is bigger or because building is what he likes to do. One of them needs the portfolio to start writing checks this spring. The other will not touch it for twenty years. Those are different portfolios, and the difference starts with the amount of time, and therefore risk, that is allowed into the portfolio.

A basic lesson on time and money belongs here. The longer you have, the more risk you can take. Not speculation, but risk in the sense that matters: the ability to sit through volatility in pursuit of the highest long-term return. In plain terms, it is the difference between owning high-quality stocks, or funds of them, and owning cash, money markets, or short-term government bonds. Over the long run, stocks have earned more than bills. Most people know that, so it gets priced in. Stocks trade at a premium for what they are expected to earn, and that premium is what can fall. So in the short run, nobody knows what stocks will do. Even when you buy the way we love to buy, with a margin of safety, you cannot fool yourself into thinking the market will agree with you, and especially not on your timeline. The market can be irrational a lot longer than you can wait it out.

That cuts both ways. Beyond the Status Quo, by Anarkulova, Cederburg, and O’Doherty, tested more than a century of returns from dozens of countries and found that a portfolio of only stocks left retirees with more wealth and a lower chance of running out than the standard mix of stocks and bonds. It also fell much further along the way, and the result depends on holding through those drops while drawing income. The conclusion to take from this is:

1. Money with time goes long, for the return.
2. Money with no time, that will be needed in the short term, is kept in safer, lower-risk assets.
3. The more time, the further up the risk-return spectrum you can go.

So the money divides in two: short-term money, which has to be there when you spend it, and long-term money, which has time. How much of each is the first real question, and there is no standard percentage. Would you be comfortable, in a bad year, with three to six months of your money in safer, slower investments? Or would you want two years of safety to sleep well, or three, or five? None of those answers are wrong. And you do not have to answer directly, or today. This is usually something that takes time to develop an answer for, and even then, it will change. We start by developing this together, and we revise it over time.

Short-term money sits in things that are very unlikely to fall by half: cash, Treasury bills, money markets, CDs, and ladders of high-quality bonds and municipal bonds of shorter duration. Every holding in the portfolio has a job. When the job is safety first and yield second, we go shorter and safer.

Holding years of spending in safer assets does three things at once.

1. It costs return. Over time, safer assets earn less than stocks, and that drag is real.
2. It pays the risk-free rate.* Cash and Treasury bills pay the rate the investment world measures every other asset against, and they pay it in years when nothing else does.
3. It is dry powder. A reserve of safer money is a put option on the market with no expiration date. It grows while it waits, and it gets used when better opportunities appear than you can find today, which is usually when the market is falling. Having cash when no one else does is powerful. It is a core rule of value investing and of how Buffett has run his own company. It is also a habit you see more often in investors who have lived through a few bear markets than in those who have only seen a bull market.

The risk of not holding enough in short-term reserves, of having to draw down the longer-term, higher-returning equity side of the ledger during a drawdown, has a name. Sequence-of-returns risk is the risk of being forced to sell a good long-term asset during a temporary decline, which turns a loss that would have recovered into one that never does and gives up the years of growth that would have followed. It does the most damage early in retirement, when withdrawals are largest relative to the years of compounding still ahead.

The person, and the return that comes from staying

A portfolio only delivers its return to an investor who holds it through the full cycle. Morningstar’s annual Mind the Gap study puts the shortfall between what U.S. funds returned and what their investors actually earned at about 1.2 percentage points a year over the decade through 2024, with the widest gaps in the funds whose investors traded most. How much of that gap is timing is debated in the literature, but the direction is not. The return a family actually earns is bounded by the strategy they can stay in.

Michael Kitces, one of the planning writers we read most, separates what people call “risk tolerance” into three things. Capacity is objective: how much loss the plan can absorb, given the time horizon, the income the portfolio has to produce, and what else the family owns. Tolerance is the willingness to accept a bad year for a better decade; it is a fairly stable trait, and a well-built questionnaire measures it reasonably well. Perception is how risky the market feels right now, and perception is the one that swings. A bear market does not change tolerance. It changes perception. What a questionnaire cannot tell us is capacity, which takes the plan, or what a person will actually do, which takes the history. What they did the last time the market fell. What money meant in the house they grew up in. What they lost once and have not forgotten. Those come out over a few conversations, and they are what we are managing when the market gets loud.

Howard Marks describes investor psychology as a pendulum that swings between euphoria and despair and spends almost no time in the middle. That swing is perception, and it is why the same person buys near the top and sells near the bottom while believing, both times, that they are being careful.

Two schools of thought

Portfolio construction has two traditional schools of thought, and each answers a question the other cannot.

1. Modern portfolio theory is about how the pieces are arranged. Stripped of its math, it comes down to a few disciplines. Don’t bet everything on one thing. Mix holdings that do not all move together. Expect more return to come with bigger swings. Set a mix, write it down, and rebalance back to it, so the portfolio does not drift into one that amplifies bad years. Its strongest argument is William Sharpe’s: before costs, the average actively managed dollar earns what the market earns, and after costs it earns less. So beating the market means being better than the other people trying, after what it costs to try. Its limits are real. It measures risk as how much prices bounce, which is not the same as losing money for good. It assumes holdings keep behaving the way they used to, and in a crash nearly everything moves together. And it has nothing to say about what any asset is actually worth.
2. Fundamental analysis is about what the pieces are. It starts from one idea: every asset is a claim on something real. A share of stock is a piece of a business. A fund is a basket of holdings, a cost, and a manager or set of rules making the decisions. A bond is a promise from a borrower. Risk, in this school, is the chance of losing money for good, and the way to avoid it is to know what you own. It is comfortable concentrating in what it understands and uncomfortable owning a hundred things it does not. Its limits are the mirror image. You can be wrong about a business or a manager, knowing either takes real work, and a concentrated portfolio that is wrong loses real money.

Side by side, neither school decides how much risk goes into a portfolio. The person and the plan decide that. Modern portfolio theory attempts to teach how risk should be spread. Fundamentals decide what the portfolio actually owns. A portfolio built on the theory alone owns things nobody has looked at closely. A portfolio built on fundamentals alone can be right about every holding and still be arranged so that one bad year undoes it.

Used together: the person and the plan decide how much risk, modern portfolio theory gives a starting point for the shape to discuss and work from, and fundamentals decide the contents. What to actually own, and how much of it, comes from knowing the asset.

Modern portfolio theory is the study of how to combine assets so that the whole carries less risk than its parts, and how to keep it that way. Its guidance on how spread out the long-term money should be, and when to rebalance, comes from seventy years of testing on how diversification and rebalancing behave across every kind of market. That work began with Harry Markowitz’s 1952 paper Portfolio Selection, which later earned him the Nobel Prize. Drawdowns are normal, they always come, and they never arrive on schedule. This is how a portfolio stays ready for them without anyone having to see them coming.

Four common ways to structure a portfolio

Advisers, family offices, and investors structure and talk about diversified portfolios in a handful of common ways, and families often arrive speaking the language of one of them. Not all of these grew out of modern portfolio theory. Buckets come from behavioral finance, and guardrails are a spending rule rather than a portfolio theory. But each one leans on the two disciplines the theory formalized, diversification and rebalancing, and these four are the ones most families have heard of.

Retirement buckets have a labeled bucket for each purpose: a cash bucket for the next year or two of spending, an income bucket of bonds for the several years after that, and a growth bucket of stocks for everything further out. Guardrails set a spending rule: take a little more when the portfolio is ahead of plan and a little less when it is behind, decided by a formula rather than a mood. Sixty-forty is one mix, sixty percent stocks and forty percent bonds, held for decades and rebalanced back to it. The all-weather allocation is a mix built to hold up in every kind of economy, growth or recession, inflation or deflation, and then left alone. Underneath, they rest on the same diversified, rebalanced portfolio. What differs is which one a person believes in enough to hold onto through a bad year.

Three approaches, side by side

Set the simplest option beside the two schools and there are three broad approaches. Each has strengths and weaknesses worth being plain about.

Each of the three above has strengths and weaknesses. The first two are easy and a good way to live with money, and over long periods most people who take them end up ahead of most people who do not. History says most active investors on the right side of the spectrum who claim to beat the market either did it over a short stretch or did not actually beat it. The SPIVA scorecards from S&P Dow Jones Indices, kept for twenty-five years, find that over any fifteen-year period roughly nine in ten actively managed U.S. large-cap funds trail the index, and that the ones that win in one stretch rarely win the next. The research on the few that do beat it over a decade or more, notably Cremers and Pareek in the Journal of Financial Economics, finds two traits. Their portfolios look very different from the index, and they hold what they own for a long time. The rest get caught when the part of the cycle they did not plan for arrives.

Buffett’s line is that you find out who has been swimming naked when the tide goes out. Everyone does well in a bull market. You do not find out who owns real quality until the market gets tough. The arithmetic underneath is unforgiving. Lose half and you need to double to get back to even, so the highest returns over a lifetime tend to belong to the people who never took the big loss. Concentration done well earns more than either of the others. Concentration done poorly gives back years at once. You only find out which one you did over time, when the market, which you had to believe was irrational in order to concentrate at all, either eventually agrees with you or not.

Fundamental analysis: knowing what you own

Fundamental analysis is knowing what you own, and that covers more than any one article can hold. Here are four of the questions that matter most, and each has its own pillar in this Library. For a business, they run like this. What the business earns on the capital in it, and on the next dollar it invests. What advantage lets it keep earning that, and how close the competitors are. How the people running it handle the cash, and if they think like owners. And what the price already assumes. That last one is an idea I think Michael Mauboussin communicated best, with Alfred Rappaport, in their book Expectations Investing. A stock price is a set of expectations about the business, priced in today. The opportunity shows up when the business turns out to be something other than what the price assumed. Most of the time it turns out to be about what the price assumed. For a fund, the questions are the same ones turned on the fund. What it actually holds. What it costs every year. How often it trades. What the manager or the rule behind it did through a bad stretch.

Great investment thinkers on the subject: what they did, and what they told everyone else to do

Three investors that really influenced our thinking told the public one thing and did something else with their own money, and both the advice and the practice were honest.

Buffett ran his company for six decades with a handful of positions producing most of the return, and in some years a single company was a large share of the whole. He has said that spreading money across many holdings is protection against not knowing what you are doing. His will, though, puts most of his wife’s money in a low-cost index fund and the rest in short-term government bonds. He explained the gap himself: “There is nothing wrong with a ‘know nothing’ investor who realizes it. The problem is when you are a ‘know nothing’ investor but you think you know something.” The will comes from the thing he understands better than almost anyone. Compounding only works if it is never interrupted. The surest way to interrupt it is to think you know what you are doing when you do not. He knows, and he has the temperament for it. He assumes his readers may not, and he would rather they do well than imitate him.

Peter Lynch ran Magellan with more than a thousand stocks at times, which sounds like the opposite of conviction until you notice that his biggest ideas were sized many times larger than the rest. What he told regular people was to own a handful of businesses they could actually follow, in industries they understood, and to hold them. He warned against selling the ones that were working to keep the ones that were not, and called that pulling the flowers and watering the weeds.

Michael Mauboussin has not run a concentrated public portfolio. He worked out the arithmetic underneath the other two. A holding should be as large as your real edge justifies and no larger, and most people think their edge is bigger than it is. He also named the paradox of skill: as investors as a group get better, luck decides more of the outcome and the average gets harder to beat. For someone without a real edge, that is a strong argument for owning the market and leaving it alone.

What the long-term money should hold

No two portfolios have the same shape, because the person and the plan decide it. What the long-term money holds turns on a few more things about the person, each of which has its own longer treatment elsewhere in this Library.

1. Knowledge, ours and yours. Concentration is grown into by knowing what you own, business or fund, and confidence alone does not get you there. Where the research has been done and there is a view, the long-term money can hold particular businesses or funds, sized by how well they are known. Where it has not, it holds the market.
2. Temperament. Some people want a simple portfolio and should have one. Others save more, invest more willingly, and hold on better when they own things they have studied and believe in. I will say that this is my own temperament. I love allocating capital to places I believe in after researching them, and then combining that with the challenge of not being emotional about it and continuing to have good judgment.
3. The rest of the balance sheet, and where it is headed. A family whose business is worth more than the portfolio already has a concentrated position, and the portfolio should be built around that fact, influenced by it and their experience from it, not on top of it. That experience can send a family in either direction: some want the most durable, most diversified option there is for everything outside the business, and some have grown comfortable with concentration and want to own five or six businesses they understand as well as their own. Both are reasonable.
4. Time. A family drawing income from the long part of the portfolio has less room for downside than one with all of their income and near-term liquidity needs met by the safer, shorter-term side of the portfolio.
5. Tax. At the top brackets, and especially in California, the construction can bend toward tax-advantaged holdings, and tax-equivalent yields are calculated with the whole financial plan in view before any security is chosen or placed. That is a large part of why a comprehensive plan, and a CFP® building it, comes before security selection. The Tax pillar spends a whole section inside the portfolio.

None of these is a rule that decides it for you. They are the things we talk through, and the answer is allowed to be different in five years.

Keeping it in shape

A portfolio drifts. Long-term money grows, short-term money gets spent, and a mix set carefully three years ago is a different mix today. There are three ways to handle that, and most portfolios use more than one.

1. Tolerance bands. Each holding, and the mix as a whole, is given a range it is allowed to drift within. When it crosses the edge, it is brought back, and some of what has grown moves over to refill the short-term reserve. Kitces’s research supports bands over calendar dates: they trade only when there is something to fix, and rebalancing usually costs a little return in exchange for keeping the mix where the plan put it. When stocks have dropped, the reserve does the job it was built for: spending comes out of it while prices recover, and whether anything gets sold to buy what fell depends on how many years it holds.
2. Events. Some changes are made because life did something: a business sold, a year of unusually high income, a large gift, a retirement. Others because the asset did: the reason for owning it changed, or a better use for the money appeared.
3. Letting it ride. For some holdings and some people, the right answer to drift is to do nothing. Let your winners run, as they say, and there is research behind the saying: rebalancing usually lowers long-term returns even as it lowers risk, and the funds that beat the market over long periods are the ones that hold what they own. A business that has grown into a big position because it kept doing exactly what you bought it for is the whole point of owning it. Trimming it back to a target because a spreadsheet says so is how people sell their best idea early. The line was drawn to be moved, and the holding has earned it.

Whichever of these applies, every so often the first questions get asked again, because the person has moved too. Is the goal still the goal? Is three years of safety still enough to let you sleep well at night without worrying about getting through a drawdown? Did the last downturn teach you something about yourself, and about what you are comfortable or not comfortable with in your portfolio?

Where this lands

How much risk a portfolio carries is decided by the person and the plan. Its shape comes from the theory and the approaches built beside it. What it owns comes from knowing the asset, and how concentrated it gets is earned by knowledge and temperament. Which is why the same firm can build one family a single index fund and another family six businesses it knows cold, and both are right for them. Buffett and Munger were not contradicting themselves, and neither are we. A portfolio fits the person holding it, and people change, so we build it together and keep revising it together for as long as it has a job to do.