A practical guide to separating long-term investments from speculative positions, setting portfolio boundaries, and using a one-way ratchet that lets successful experiments strengthen the core without making the core rescue unsuccessful ones.
There is an appealing simplicity to saying that 85% of a portfolio will be invested for the long term while 15% will be reserved for individual ideas.
The difficulty begins approximately five minutes later.
What happens when one of the individual stocks doubles? What happens when several of them fall? Does the investor restore the original percentages in both cases? Can new money refill the speculative account? When does a successful position stop being an experiment? And what prevents a temporary 15% allowance from slowly becoming half the portfolio?
These questions matter because an allocation is not yet a system. An allocation describes how a portfolio looks on one day. A system describes what happens next.
In A Budget for Speculation: Can Stock Picking and Passive Investing Coexist?, I argued that some investors may be better served by giving speculation a limited budget than by pretending they will never feel tempted to speculate. Most of the portfolio performs the sober work of long-term wealth building. A smaller portion provides room for individual ideas, intellectual curiosity and the occasional attempt to be right in a more interesting way.
The behavioral premise was simple: the appetite gets a plate, not the pantry.
This companion article is about building the plate.
Core–satellite investing is a broad family of portfolio designs. The core generally consists of diversified, long-term holdings. The satellites are smaller positions used to pursue active ideas, specialized exposures or other objectives not fully represented in the core.
Nothing in the general definition requires the satellite to contain short-term trades. Nothing requires profits to move in only one direction. In an ordinary core–satellite portfolio, the investor might simply rebalance everything toward the original proportions. If the satellite grows too large, money moves from satellite to core. If it becomes too small, money may move from core to satellite.
That is conventional rebalancing: restore the target.
The system developed here is intentionally more asymmetric. It is designed for someone who views the satellite partly as a behavioral allowance for speculation and wants the long-term portfolio protected from that activity.
Under this version:
This is not the only way to manage a core–satellite portfolio. It is a particular variant we might call the one-way ratchet.
The core is not merely the largest investment on the brokerage screen. It is the part of the portfolio expected to carry the investor toward long-term goals even if every satellite idea proves disappointing.
That means its design comes first.
For one person, the core might be a broad global stock fund. For another, it might include domestic and international stock funds, bonds and other diversified holdings. Time horizon, financial circumstances and tolerance for loss matter more than any universal list of ticker symbols. Asset allocation and diversification are related but different decisions: allocation determines how much goes into broad asset classes, while diversification spreads risk within and among them. FINRA’s overview of asset allocation and diversification explains that distinction in more detail.
Emergency savings should not be counted as the core. Neither should money needed for an approaching bill, a home repair or another short-term obligation. The system begins only after money has genuinely become investable.
The most useful test is severe but clarifying:
If the satellite went to zero, would the remaining portfolio still constitute a reasonable long-term investment plan?
If the answer is no, the core is not yet carrying the whole job assigned to it.
Suppose the total investable portfolio is \(P\), and the investor chooses a satellite target of \(t\). The starting satellite allocation is:
\[ S_0 = tP \]
The starting core is:
\[ C_0 = (1-t)P \]
If the portfolio contains USD 100,000 and the chosen target is 15%, the initial structure is therefore:
The 15% figure is illustrative, not a recommendation. The important decision is not finding a mathematically sacred percentage. It is deciding, in advance, how much of the portfolio may be exposed to concentrated judgment.
A target alone, however, creates a temptation to trade every small deviation. It can be useful to choose an upper boundary as well. The investor might begin at a target of 15% but take no rebalancing action until the satellite exceeds 20% of the total portfolio.
This creates two separate numbers:
The target says where the satellite belongs. The ceiling says when its success has begun changing the risk structure of the whole portfolio.
A boundary is only useful if investments cannot wander across it whenever convenient.
Before purchasing anything, assign it a job. A broadly diversified holding chosen as part of the long-term asset allocation belongs to the core. A concentrated position based on a company thesis, sector forecast, valuation judgment or tactical idea belongs to the satellite.
The classification should not change merely because the investment has performed well.
This matters because successful speculation has a way of applying for retroactive citizenship. A stock bought as a risky experiment rises sharply, and the investor begins describing it as a permanent long-term holding. That may eventually become a defensible conclusion, but the price increase itself is not evidence that the position has become diversified or that its risk has disappeared.
If a satellite holding is to graduate into the core, it should have to satisfy the core’s rules. In most portfolios, one company cannot do that by itself. The more natural form of graduation is to sell part of the successful position and transfer the proceeds into the diversified holdings already doing the core’s work.
It also helps to place a limit on any single satellite position. If the satellite is 15% of the portfolio and one company consumes the entire satellite, the investor has bounded the size of the gamble but not diversified the experiment. A rule might cap each position at a chosen percentage of the total portfolio or at a chosen fraction of the satellite. The appropriate number depends on the investor, but the number should be selected before enthusiasm arrives.
The defining rule of the ratchet is not that money constantly moves from satellite to core. Constant movement could produce unnecessary trading, taxes and attention. The rule is that, when capital does cross the wall, it is not automatically sent back in the opposite direction.
At a review date, let:
The satellite’s current weight is:
\[ w = \frac{S}{P} \]
If \(w\) exceeds the chosen ceiling, the amount needed to return the satellite to its target is:
\[ x = S - tP \]
The investor transfers \(x\) from the satellite to the core. Because this is a transfer within the portfolio, total portfolio value \(P\) does not change.
Consider a portfolio that began with USD 85,000 in the core and USD 15,000 in the satellite. Later, the core has grown to USD 93,500 and the satellite has grown to USD 30,000.
The total portfolio is now USD 123,500, and the satellite represents:
\[ \frac{30{,}000}{123{,}500} \approx 24.3% \]
If the target is 15% and the ceiling is 20%, the ceiling has been crossed. The desired satellite value is:
\[ 0.15 \times 123{,}500 = 18{,}525 \]
The amount transferred into the core is therefore:
\[ 30{,}000 - 18{,}525 = 11{,}475 \]
After the transfer:
The satellite remains large enough for future ideas, but USD 11,475 of concentrated success has changed jobs. It now belongs to the diversified compounding engine.
This is where the ratchet becomes meaningfully different from ordinary rebalancing.
Suppose that, at another point, the core is worth USD 110,000 while the satellite has fallen to USD 6,000. The complete portfolio is worth USD 116,000, so the satellite now represents approximately 5.2%.
A conventional 15% rebalancing rule would calculate a desired satellite value of:
\[ 0.15 \times 116{,}000 = 17{,}400 \]
It would then transfer USD 11,400 from the core into the satellite.
The one-way ratchet does not do that.
The USD 6,000 satellite may remain at USD 6,000. Its shrinking share of the portfolio records the fact that the experiments have not earned more responsibility. The investor may continue operating it at the smaller size, pause it or close it. What the investor does not do is sell diversified long-term holdings merely to restore the speculative allowance.
There are at least two defensible policies for future contributions:
Either policy can preserve the wall. What weakens the system is deciding after a loss that a particularly exciting new opportunity deserves an emergency transfer from the core.
A mechanical instruction to sweep trading profits every week or sell every position after a fixed gain sounds disciplined, but it can create activity without improving the portfolio.
A stock that rises 10% has not necessarily become overpriced. A stock that has not risen may still have become a worse investment. Price movement alone cannot determine whether the original thesis remains intact.
It helps to separate two kinds of decisions:
A position might be sold because its thesis has failed, its expected reward no longer justifies its risk or its size has become excessive. The satellite as a whole might be trimmed because it has crossed the portfolio ceiling. Those are different judgments.
Reviews can occur on a modest calendar—perhaps quarterly or annually—with an additional check when a major price movement is known to have pushed the satellite past its ceiling. Threshold-based rebalancing responds to allocation drift, while calendar-based review reduces the burden of constant monitoring. Vanguard describes both approaches in its overview of portfolio rebalancing.
The goal is not to observe the portfolio as frequently as possible. The goal is to make the agreed rules difficult to evade.
A satellite budget limits risk only if the investments inside it cannot create obligations larger than the money assigned to them.
Ordinary unleveraged shares can fall to zero, but they cannot fall below zero. Borrowing on margin, selling certain uncovered options or using other leveraged structures can produce losses or forced sales that reach beyond the nominal satellite balance. At that point, the wall is decorative.
A simple governance rule is therefore:
No satellite position should be capable of forcing capital out of the core.
That does not prescribe a trading strategy. It establishes an architectural requirement. Before an instrument belongs in the satellite, the investor should understand its maximum possible loss, financing requirements and circumstances under which a broker could liquidate other holdings.
Account structure matters as well. Selling investments in a taxable account can generate capital gains or losses, and net short-term capital gains are generally taxed as ordinary income under current U.S. federal rules. IRS Topic 409 summarizes the basic treatment. Trading inside a retirement account changes the tax picture, but it also places speculative activity inside limited tax-advantaged space and generally removes the ordinary taxable-account benefit of realizing investment losses. Taxes should therefore be part of the design, not an afterthought or a universal reason to put the satellite in one particular account.
A bounded experiment can still be an expensive experiment.
The proper comparison is not whether the satellite made money. In a rising market, many strategies make money. The more revealing question is whether the satellite performed better than the same money would have performed in the core.
For a satellite with no later deposits or withdrawals, the comparison is straightforward. If the initial satellite value was \(S_0\) and the core benchmark returned \(r_c\) over the period, its hypothetical benchmark value is:
\[ B = S_0(1+r_c) \]
If USD 15,000 in the satellite becomes USD 18,000 while the equivalent core investment would have become USD 19,200, the satellite produced a positive return and still underperformed its alternative by USD 1,200.
With later contributions, withdrawals and transfers, the accounting becomes more involved. A practical ledger can record each cash flow and compare it with a hypothetical purchase or sale of the chosen core benchmark on the same date. At minimum, the scoreboard should include:
This comparison should be made over years, not a few fortunate months. Its purpose is not to shame the investor into abandoning an activity they enjoy. It is to reveal the price of that enjoyment—and whether genuine skill is appearing strongly enough to overcome it.
An investor may reasonably conclude that modest underperformance is an acceptable entertainment cost. That is a personal judgment. But it should be made with the cost visible.
The rules do not need to become a fifty-page investment policy statement. A useful version could fit on one screen:
Core: The complete long-term portfolio, designed to remain viable if the satellite is lost.
Satellite target: 15% of the investable portfolio.
Satellite ceiling: 20% of the investable portfolio.
Single-position limit: A predetermined share of the total portfolio or satellite.
Transfer rule: If the satellite exceeds its ceiling at a review, trim it back to the target and move the proceeds into the core.
Loss rule: Never sell the core merely to restore a diminished satellite.
New-money rule: State in advance whether the satellite is closed-budget or may receive a fixed share of new contributions.
Risk rule: No borrowing, undefined-loss position or other instrument capable of forcing capital out of the core.
Review schedule: Review on predetermined dates and when the ceiling is clearly breached—not whenever the market becomes exciting.
Scoreboard: Compare all satellite results with the same cash flows invested in the core.
The percentages above are examples. The value lies in deciding what the rules are before fear, confidence, boredom or a spectacular investment story begins negotiating with them.
The satellite will test the system in two opposite ways.
When it succeeds, the investor will want to believe the ceiling is unnecessarily cautious. When it fails, the investor will want to believe the next idea deserves fresh capital from somewhere else.
The one-way ratchet refuses both invitations.
It does not interpret a winning streak as permission for speculation to annex the portfolio. It does not interpret a losing streak as an obligation for the diversified core to finance a comeback. Success may send bricks across the courtyard. Failure remains inside the workshop.
That asymmetry is the point.
A core–satellite allocation can be drawn as two circles on a page. A functioning portfolio requires something more: definitions, ceilings, transfer rules, a scoreboard and a decision about what happens after disappointment.
Build those while the market is quiet. Then, when excitement arrives, the portfolio will already know what to do.
This article discusses portfolio design for educational purposes and does not recommend a particular allocation, security, account type or trading strategy. Investments can lose value, including diversified funds. Tax treatment and an appropriate portfolio structure depend on individual circumstances, goals, time horizon and tolerance for risk.
Facebook
Reddit
WhatsApp
Twitter
Kommentare
0 Kommentare
Melde dich an, um am Gesprach teilzunehmen.
Noch hat niemand geantwortet. Kommentare erscheinen hier, sobald das Gesprach beginnt.