Leeway Research

Portfolio strategy 17 min

The portfolio without a sell button

Single stocks, 40 years, not one sale: the plan can work. But the outcome rests on the entry price, the cash coming in and the tax home, not on the stocks picked in 2026.

The assessment is produced by a discussion among several models, with continuous fact-checking and research. Jump to the method

An investor who buys shares and never sells them saves fees, taxes and many of their own mistakes. They also give up the most important tool any asset manager has: reallocation. This analysis tests whether a portfolio of single stocks only, with no funds, bonds or commodities, can last 25 to 40 years without a single sale. The answer is yes, on four conditions.

The arithmetic at a glance

Against a low-cost global index the edge is small: 0 to 1% a year, with wide dispersion. Against expensive active funds it is clear. It depends on four things the investor controls.

  • The entry price. If you never sell, you keep your purchase price forever. Entering at 16 to 18 times earnings protects more than any clever stock pick.
  • Breadth. The portfolio holds 50 to 80 companies in roughly equal weights. No region gets more than about 35% at the start.
  • A steady inflow of cash. Dividends, takeover proceeds and new contributions replace reallocation. They go wherever new money earns the most.
  • A rulebook before the first purchase. It sets out what happens with rights issues, takeovers, tax bills and inheritance. Otherwise everyday life forces exactly the sales the approach forbids.

0 to 1%

That is a realistic annual edge over a low-cost global index. Against active funds it is 1 to 2.5% after costs, depending on the tax home.

7.5 to 9%

This is the nominal annual return to expect from an entry at 16 to 18 times earnings. After inflation about 4 to 5% remains, before taxes in the investor's home country.

At most 2%

Short positions stay a small, separately held sleeve with a fixed loss limit. Their expected contribution is zero.

The Portfolio at a Glance

The portfolio comprises 27 positions across 6 building blocks with 23 long and 4 short positions.

The following overview summarizes the strategic building blocks, position directions, and allocated assets for this research theme.

Long

Companies that reallocate their own capital

9 assets

These companies exit old businesses and reinvest without the shareholder having to sell. For a portfolio without sales that is the most valuable trait. Partners Group adds exposure to value created before companies list. Together they get 20 to 25% of the portfolio.

+6
Long

Industrial bottlenecks, bought on setbacks

4 assets

Chip manufacturing, lithography, wafers and industrial gases are hard to replace, but cyclical and often expensive. They are bought only when order cuts push prices down. The share is 15 to 20%.

+1
Long

Networks, exchanges and mineral rights

5 assets

Rail networks, exchanges and royalty rights on resources collect recurring fees and feed renewal with dividends. They offer no protection in a crisis. The share is 15 to 20%.

+2
Long

Quality after the fall

1 assets

Strong healthcare and consumer brands enter the portfolio only at a price that not everyone is already paying. Novo Nordisk stands for this pattern. The building block gets 10 to 15% in total.

Long

Banks and platforms in growth markets

4 assets

Emerging markets will supply many of the new listings and new investors. The portfolio buys leading companies in countries with reliable legal systems. The share is 10 to 15%.

+1
Short

A small short sleeve against chronic dilution

4 assets

These positions are optional, held separately and together no larger than 2% gross. They bet on falling value per share at companies that keep funding themselves through new shares. They are not a hedge for the core portfolio.

+1

The analysis

The brief sounds simple. One portfolio, single stocks only: no funds, no ETFs, no bonds, no commodities. It is bought now and held for at least 25 years, preferably 40. Nothing is ever sold. Most of it is long, with a few targeted short positions.

Think it through and an uncomfortable insight follows quickly. The clever picks of 2026 matter little. In 25 years most of today's names will have been acquired, delisted or folded into successor companies. The outcome rests on the price paid at entry, on breadth, on the cash that arrives later and on the rules for investing it.

The entry is not cheap. At the end of October 2026 the MSCI ACWI world index traded at about 21 times trailing twelve-month earnings.1 The S&P 500 stood at 23.5.2 A portfolio that wants to enter at an average of 16 to 18 times earnings therefore has to buy where valuations are not yet stretched: in Japan, in Europe and in selected emerging markets, in quality stocks after a sell-off and in industrial stocks during an order slump.

From that entry the return can be roughly derived. The base is the earnings yield, meaning earnings per share after necessary investment, relative to the price. Growth per share comes on top. A slightly lower valuation level and the friction of withholding taxes are deducted. Against a low-cost global index the portfolio ends up close to parity. It beats expensive active funds by 1 to 2.5 percentage points a year after costs, and bonds and cash by a wide margin in real terms.

What the expected return is made of

Percent per year, for an entry at 16 to 18 times earnings. Each bar shows the midpoint of its range.

  1. Earnings yield after necessary investment+4.5 to 5%
  2. Growth per share, nominal+4 to 5%
  3. Slightly lower valuation level0 to −0.5%
  4. Withholding taxes and other friction−0.3 to −0.6%
  5. Expected return, nominal7.5 to 9%
  6. Inflationabout −3.5 to −4%
  7. Expected return after inflation4 to 5%

The end values are rounded cautiously; the midpoints of the individual parts add up to slightly more. Taxes in the home country and taxes on wealth are not deducted. The section on cash shows how much they matter.

How the arithmetic takes shape

No selling does not mean standing still

The owner can stay forever because the companies do not.

The obvious picture of a forever portfolio is a photograph: you buy a selection in 2026 and freeze it. That picture does not hold for long. The ban on selling works more like a valve that opens in one direction only. Buying remains allowed. Dividends, cash from takeovers and new contributions can be invested at any time.

The comparison with an index sharpens the view. A market-cap-weighted index never sells its winners either. Its real advantage lies elsewhere: it keeps adding new companies. That matters more than it sounds. A study of almost 64,000 stocks worldwide shows that between 1990 and 2020 just 2.4% of companies generated all the wealth created above the return on short-dated US Treasury bills.3 An investor who does not already own the future winners has to be able to buy them later.

That shifts the question. What matters is not which stocks sit in the portfolio in 2026. What matters is how much cash arrives over the following decades and by which rule it is invested.

Cash sets the limit, not stock selection

A portfolio without sales lives on its inflows. On one side are dividends after withholding tax, cash from takeovers and new contributions. On the other are taxes on wealth or on unrealised gains, withdrawals for living costs, money for rights issues and administration. If inflows cannot cover the unavoidable outflows, the approach is not just risky. It is unworkable, because then the tax bill forces the sale.

Our first estimate was too pessimistic. Calculated at statutory withholding rates, a gross dividend yield of 2.5 to 3% shrinks to 1.2 to 1.7%. But private investors usually pay the lower treaty rates, often 15%. On Japanese dividends non-residents pay 15.315%, not 20.315%. With reclaims and credits, about 2 to 2.5% remains before tax in the home country.4 So the concern about withholding tax is justified, but its size was overstated. Reclaims remain a nuisance. Switzerland initially withholds 35% from foreign investors, Denmark 27% and Germany 26.375%, and recovering the excess takes time. Taiwan has few treaties.

The tax home weighs most, more than any single stock idea. Where the state taxes wealth itself, for example through Switzerland's cantonal wealth tax of typically 0.1 to 1%,5 cash gets tight. Where it taxes unrealised gains, as the planned reform of the Dutch Box 3 regime intends from 2028,6 the arithmetic tips over. German residents have an advantage that the fund ban even increases: single stocks carry no Vorabpauschale, the advance lump-sum tax on funds. Only dividends and sales are taxed.

The same dividend in three countries

What remains for new purchases per 100 euros or francs of portfolio value each year, at a 2.75% gross dividend yield and an 8% expected total return.

Netherlands

Planned Box 3 reform from 2028: 36% on the actual return, including gains on shares not sold.

Gross dividend
2.75%
Withholding tax that never comes back
−0.10%
Domestic tax on the dividend
included in Box 3
Tax on holding
−2.88%
Left for new purchases
−0.23%

The tax exceeds the dividend. Without new money, sales become compulsory.

Switzerland

Dividends taxed as income, plus cantonal wealth tax. Private capital gains stay tax-free.

Gross dividend
2.75%
Withholding tax that never comes back
−0.10%
Domestic tax on the dividend
−0.83%
Tax on holding
−0.50%
Left for new purchases
1.32%

There is enough for new purchases, but the wealth tax costs almost a fifth of the gross dividend.

Germany

Flat tax on dividends and sales, no wealth tax, no Vorabpauschale on single stocks.

Gross dividend
2.75%
Withholding tax that never comes back
−0.10%
Domestic tax on the dividend
−0.73%
Tax on holding
none
Left for new purchases
1.92%

Most is left for new purchases. It only gets expensive when a takeover forces a sale.

Simplified calculation, not tax advice. Assumptions: in Switzerland a 30% marginal rate on dividends and a 0.5% wealth tax; in Germany 26.375% including the solidarity surcharge, with foreign withholding credited; in the Netherlands 36% on an 8% total return. Church tax, allowances and loss carry-forwards are ignored. In takeovers the picture partly reverses: Germany then taxes the capital gain, while Switzerland leaves it tax-free.

The entry price stays forever

An investor who may sell can correct an overpriced purchase later. One who never sells keeps the entry price for the whole holding period. The history of the Nifty Fifty shows what that means. The celebrated American growth stocks of the early 1970s cost on average 42 times their earnings, against 19 times for the S&P 500. The companies did go on to grow faster than the market. Even so, the basket held unchanged returned 11.2% a year to the end of 2001, slightly less than the 12.0% of the broad market.7 The growth was correctly identified and overpaid for.

Over 40 years, however, a falling valuation level loses much of its sting, as the table shows. In the end what counts most is how much cash a company generates per share, after necessary investment and after dilution from new shares.

Two conclusions follow. First, today's record margins at American companies are priced in twice. A multiple in the high twenties rests on profit shares that were carried by low taxes, cheap money and open world markets. A state running large deficits will take part of that back through taxes, price regulation and special levies. That is a political risk, not a cyclical dip.

Second, regional leadership rotates. Japan accounted for about 45% of world stock market value in 1989. That is why no region gets more than about 35% of the portfolio at the start. With the US at 60 to 65% of the global index, this is a deliberate underweight. It rests on the permanent entry price and on political risk, not on an expectation that the US will lose.

What a falling P/E costs each year

The table shows the annual return lost when the valuation level falls evenly from entry to the end of the holding period. All values refer to the price-to-earnings ratio on trailing twelve-month earnings.

P/E at entry and at the endHolding periodReturn lost per year
From 30 times to 20 times25 years−1.6 percentage points
From 30 times to 20 times40 years−1.0 percentage points
From 50 times to 20 times25 years−3.6 percentage points
US-heavy global index: from the high twenties to about 20 times25 to 40 years−0.7 to −1.6 percentage points
This portfolio: entry at 16 to 18 times, similar at the end25 to 40 yearsclose to zero

Over 40 years an expensive entry is not a death sentence. It is, however, a permanent burden that can no longer be corrected. For capital-intensive companies such as TSMC or ASML, what counts is therefore less the reported P/E than the cash left after all investment.

New money has to compete

Once inflows replace reallocation, the rule for using them becomes the portfolio's most important lever. Two simple rules suggest themselves, and both lead astray. Steering new money automatically into the smallest positions systematically buys companies whose decline has not yet reached their cash flow. Never topping up anything that has fallen starves every recovery.

The better question is: where does new money earn the most? The test covers demand for the products, customer loyalty and competitive position, plus investment needs and dilution from new shares. In each investment round the existing holdings compete against a standing slot for newcomers. Every two to three years, one to three new companies join the portfolio this way.

Dividends and buybacks also play different roles. A buyback raises the loyal shareholder's stake without triggering tax. But it only creates value if the company buys below intrinsic value. At high valuations it transfers wealth from the shareholders who stay to those who sell. Dividends, by contrast, are the fuel of renewal because they can be invested freely. Where the tax home taxes dividends heavily and spares capital gains, buybacks are the better form. Renewal then has to come from new contributions.

One weakness remains: involuntary exits arrive at the wrong time. Controlling shareholders and private equity buyers take companies private when they think them cheap. In Japan a two-thirds majority is enough. A permanent holder's cash-outs therefore cluster in weak markets, are taxable and must be reinvested. Typical takeover premiums of 20 to 30% soften this, but someone else picks the moment. Listed subsidiaries with a controlling parent therefore get a large weight only if the purchase price already reflects that risk.

Short selling and permanence do not mix well

Read strictly, opening a short position is already a sale. But even read generously, the technique fits badly into a portfolio meant for decades. Borrowed shares can be recalled. Borrowing fees, dividend compensation and forced buy-ins all run against the short seller. In the US, gains on short sales are generally short-term regardless of holding period, and shorting shares you own triggers a constructive sale. Above all, a short behaves like a mirror image of the core portfolio: a winning position shrinks and a losing one grows. And covering does not create new buying power. It only releases the collateral that was posted.

The obvious candidates would be the market's well-known problem cases, such as AMC, Plug Power or Tilray. These are exactly the ones that drop out. They sit at the centre of retail speculation that can multiply prices within days. AMC also completed a refinancing of almost $4 billion in October 2026.8 That removes the trigger a bet against the company would have relied on. Tilray faces a price jump if the US moves cannabis into a milder legal category.

What survives is one documented finding: companies that keep issuing many new shares underperform on average. But in hard-to-borrow stocks, borrowing fees and squeezes largely eat up the apparent excess return.9 If at all, then only a small, separate sleeve belongs in the portfolio: at most 2% gross and 0.3 to 0.5% per name. Candidates are companies with no dividend, a high market value relative to revenue and persistent dilution. The sleeve is opened once at the start, at a separate bank, with cash collateral and without lending out the investor's own shares. It is covered only at three times the entry price, after an 85% fall or a delisting, or when the borrowing fee has stayed above about 10% a year for three months.

The strongest counter-argument is to drop the sleeve entirely and hold a cash buffer funded by dividends instead. That is almost as good. Investors who insist on short positions keep them so small that they cannot do damage.

The rulebook comes before the first purchase

A portfolio without sales rarely fails because of a bad stock. It fails because of a letter from the custodian that nobody was prepared for. So the answers are written down before the first order goes out. Some rules even pay. In France and Italy many company statutes reward loyal, registered shareholders with extra voting rights, usually double votes after two years.10 Other rules protect against losses nobody expects. In the US, accounts whose owner stays silent for years can pass to the state, and the securities are then often sold.11

Rights issue

Rights are always taken up, or the scrip dividend is chosen where available. The money comes from the cash buffer and never from a sale.

Takeover offer

Shares are tendered only if a squeeze-out or a barely tradable remnant would otherwise follow. An attractive price alone is no reason.

Cash-out

Cash from a squeeze-out is invested by the same rule as every other inflow: wherever new money earns the most.

Dormant accounts

At least once a year the owner logs in at every custodian and confirms their details. That way no account passes to the state as abandoned.

Inheritance

Investors without US residence can owe US estate tax of up to 40% on US shares above $60,000. UK shares fall under UK inheritance tax, and Canadian residents are taxed on a deemed sale at death. This is settled before buying.

Loyalty bonuses

Air Liquide pays shareholders who keep their shares in registered form for two full calendar years a 10% higher dividend, capped at 0.5% of share capital per holder. L’Oréal offers a similar premium. It is a welcome gift, but no reason to organise custody around it.

The world this portfolio has to grow old in

A scenario for 2031 to 2046, not a forecast. It determines which companies will have to prove themselves.

For 2031 to 2046 we consider an environment of fiscal dominance the most likely. Government deficits in developed economies stay high, driven by ageing, defence, energy and interest on legacy debt. First the bond markets discipline governments, then governments respond with a leaky form of financial repression. The first signs are already visible: in October 2026 the term premium on French government bonds stood at about 2.3 percentage points.12 We give this sequence roughly 55%. Austerity or a clean, AI-driven disinflation each get 20 to 25%. Ten-year government bond yields move between 3.5 and 5.5%, and inflation runs a little higher and is more volatile than in the decade to 2020.

The state increasingly takes its share through taxes on holdings rather than transactions: wealth taxes, taxes on deemed returns, exit and inheritance taxes, and pressure on the US step-up in basis at death. On top come price caps, special levies, export controls and frozen fees at visible bottlenecks. At the same time artificial intelligence shifts competitive moats. Routine desk work and intermediaries with little capital lose value. Physical, regulated and trust-based businesses gain, and exactly these become political targets.

An unlevered, stock-only portfolio stands broadly on the right side in this world. It owns claims on nominal growth and real cash flows, the main escape from repressed bond and deposit rates. And an investor who is never forced to sell, but has inflows, can use the sell-offs that short-term traders trigger. The portfolio is exposed through its tax home, through companies at bottlenecks that face price caps or sanctions, through involuntary exits and through succession at companies shaped by their founders.

For valuations this means a decline first and a partial recovery later. By around 2031 US valuations are likely to approach 19 to 22 times earnings, and celebrated quality stocks are likely to fall from 40 to 50 times towards 24 to 30 times. Japan should stay at 13 to 16 times thanks to buybacks, and India should approach 20 times under the weight of many listings. After that, global valuations settle at 16 to 20 times. A portfolio that enters at 16 to 18 times therefore ends roughly where it started.

Four phases of unequal length

The width of each bar matches its duration. Building takes two years, passing the portfolio on takes two decades.

2051: the 25-year checkpoint. By then most of the original companies have been acquired, delisted or folded into successors.

  1. 2026 to 2028

    Building

    Tax home, inflows, withdrawals and reserve are settled first, then the portfolio is bought in tranches over 9 to 18 months. The cheapest parts come first: Japan, trading houses, emerging-market banks and healthcare stocks after their sell-off.

  2. 2028 to 2032

    Proving

    The investment in AI and power grids has to pay off, and depreciation catches up with capital spending. Expensive growth stocks lose valuation, and inflows buy the setbacks. The first verdicts come in on Berkshire under Greg Abel and on Constellation without Mark Leonard.

  3. 2032 to 2046

    Repression

    Equities become the escape from negative real rates, and taxes on holdings rise. Bottlenecks are regulated like utilities. The three to five largest positions reach 25 to 40% of the portfolio, so all new money goes elsewhere.

  4. 2046 to 2066

    Succession

    The portfolio now resembles a self-selected index of the 2026 survivors plus everything the inflows bought later. Clean estate planning now protects the most value.

In reality the phases overlap; the boundaries are working assumptions. The policies of the large states will shape the sequence more than the calendar.

The portfolio's framework

The full portfolio holds 50 to 80 companies in roughly equal weights. The 23 long positions on this page are its framework and together make up just over a third of its value. Further names from the same building blocks fill the rest, and 2 to 5% stays in cash. The same rule applies to every position: if its reason falls away, it gets no new money. It is not sold.

Leeway's ratings and this analysis are two separate judgements. The ratings assess today's business and the current valuation against the stock's own history. The analysis asks whether a company reallocates its capital wisely over decades without the shareholder having to sell.

The Business-Rating scores the quality of the business model, independent of the share price. The Market-Fit-Rating tests eighteen fundamental figures for how well the company currently fits the market; a negative reading implies expected negative performance. The Cycle-Rating places the valuation in the stock’s own history: a higher figure means the shares are historically cheaper. The Leeway-Score combines the three in equal parts. How the ratings are calculated

The recommendations since publication

From 18 November 2026 to 19 November 2026: Portfolio+1.7%ACWI-2%

Companies that reallocate their own capital

Long

These companies exit old businesses and reinvest without the shareholder having to sell. For a portfolio without sales that is the most valuable trait. Partners Group adds exposure to value created before companies list. Together they get 20 to 25% of the portfolio.

Berkshire Hathaway Inc.

BRK-B.US · Financial Services · 1.2tn USD

Long

Role in the portfolio: Berkshire buys in crises and reallocates while the shareholder sits still.

Why the stock fits: Berkshire can use its insurance float and cash to buy in market crashes without you having to sell. It buys back its own shares only below its own estimate of their value, exactly the behaviour a permanent holder needs.

When buying stops: The position gets no new money if the return on newly invested capital under Abel falls to the cost of capital, or the group is broken up on poor terms.

Position and what to watch

Position: The position is about 3% and is built up over 9 to 18 months.

What to watch: Watch how Greg Abel deploys the cash pile and at what prices Berkshire buys back its own shares.

Main risk: Capital allocation after Buffett is untested. Berkshire pays no dividend and so supplies no cash for renewal. Investors without US residence carry US estate tax risk, and sheer size is a drag.

Leeway Rating

General scores - independent of the research topic

Leeway Score42.3/100

  • Business Rating 23.0
  • Market-Fit Rating Trend+1377.9
  • Cycle Rating 26.0

Check the full analysis

Investor AB

INVE-B.ST · Financial Services · 1.3tn SEK

Long

Role in the portfolio: Investor is an investment company whose succession has held up over generations.

Why the stock fits: Investor combines a century of institutionalised capital allocation by the Wallenberg family with Nordic industrial companies. Through Patricia Industries and its stake in EQT, shareholders also gain access to value created outside the stock market.

When buying stops: The position gets no new money if family control destroys value or the discount becomes permanent.

Position and what to watch

Position: The position is about 2.5%. It is topped up when the discount to net asset value widens, which often happens as real rates rise.

What to watch: Watch the discount to net asset value, the dividend trend and the valuation of unlisted holdings.

Main risk: Sweden initially withholds 30% tax, and the reclaim to the treaty rate must be filed. The family controls the company, and the share partly behaves like a fund.

Leeway Rating

General scores - independent of the research topic

Leeway Score54.6/100

  • Business Rating 60.0
  • Market-Fit Rating Trend+877.3
  • Cycle Rating 26.5

Check the full analysis

ITOCHU Corporation

8001.TSE · Industrials · 15.5tn JPY

Long

Role in the portfolio: ITOCHU works like an internal capital market in a cheap country and keeps supplying cash for renewal.

Why the stock fits: The trading house combines disciplined capital allocation with Japan's governance reforms, buybacks and rising dividends, at a modest valuation. Because it leans more on consumer businesses than on resources, its sanctions risk is the lowest among the big trading houses.

When buying stops: The position gets no new money if a commodity slump coincides with a loss of capital discipline.

Position and what to watch

Position: The position is about 2% and is preferably bought in the first purchase tranche.

What to watch: Watch the unwinding of cross-shareholdings, buyback policy and the trend in dividends per share.

Main risk: Foreign money sells Japan first in crises and when yen-funded trades unwind. Currency risk adds to that, and support from Japanese retail investors fades after 2035.

Leeway Rating

General scores - independent of the research topic

Leeway Score27.2/100

  •    
  • Market-Fit Rating 64.2
  • Cycle Rating 17.5

Check the full analysis

Mitsubishi Corporation

8058.TSE · Industrials · 17.7tn JPY

Long

Role in the portfolio: Mitsubishi earns real cash flows from resources and infrastructure and is run with discipline.

Why the stock fits: Mitsubishi is broadly diversified, holds plenty of cash and has explicitly committed to shareholder distributions. The position also spreads management risk across a second trading house.

When buying stops: The position gets no new money if the group overpays at the peak of a commodity cycle or sanctions destroy the value of its Russian stakes.

Position and what to watch

Position: The position is about 1.5%.

What to watch: Watch sanctions around Sakhalin, investment discipline at the top of the commodity cycle and the payout ratio.

Main risk: The Russian Sakhalin LNG project carries sanctions risk. Coal and LNG face structural change, and the group is complex.

Leeway Rating

General scores - independent of the research topic

Leeway Score22.0/100

  •    
  • Market-Fit Rating 61.0
  • Cycle Rating 4.9

Check the full analysis

Hitachi, Ltd.

6501.TSE · Industrials · 21.1tn JPY

Long

Role in the portfolio: Hitachi rebuilds itself and owns a bottleneck of its own in grid technology.

Why the stock fits: Hitachi sold its metals, chemicals, auto parts and construction machinery businesses and redeployed the capital. The company does this restructuring, not the shareholder. With Hitachi Energy it also owns a strong grid technology business.

When buying stops: The position gets no new money if grid orders collapse or the discipline of the group's restructuring slips.

Position and what to watch

Position: The position is about 1.5% and is bought only below roughly 25 times trailing twelve-month earnings.

What to watch: Watch order intake at Hitachi Energy and whether the group keeps selling businesses that no longer fit.

Main risk: Enthusiasm for power grids is already largely in the price. Execution risk in digital services and the yen come on top.

Leeway Rating

General scores - independent of the research topic

Leeway Score48.0/100

  • Business Rating 43.0
  • Market-Fit Rating Trend−2675.7
  • Cycle Rating 25.4

Check the full analysis

Fairfax Financial Holdings

FFH.TO · Financial Services · 55bn CAD

Long

Role in the portfolio: Fairfax uses its insurance float as leverage without the shareholder taking on debt.

Why the stock fits: Higher interest rates lift the returns on the insurer's investment portfolio. Fairfax has often bought when others were forced to sell, and it trades at a modest valuation.

When buying stops: The position gets no new money if a major reserving or catastrophe loss hits, or governance drifts after Watsa.

Position and what to watch

Position: The position is about 1.5%.

What to watch: Watch the combined ratio, reserving and succession planning.

Main risk: Its investments are concentrated in a few large bets. The company depends heavily on Prem Watsa, and Canada withholds tax.

Leeway Rating

General scores - independent of the research topic

Leeway Score46.3/100

  • Business Rating 30.0
  • Market-Fit Rating Trend−1173.2
  • Cycle Rating 35.6

Check the full analysis

Indutrade AB

INDT.ST · Industrials · 93bn SEK

Long

Role in the portfolio: Indutrade renews itself from within, so the shareholder never needs to sell.

Why the stock fits: Indutrade keeps buying small industrial niche businesses and runs them in a decentralised way. The model does not depend on a founder and is less crowded than the best-known serial acquirers.

When buying stops: The position gets no new money if organic growth stalls and acquisition prices rise.

Position and what to watch

Position: The position is about 1.5% and is bought only at a sensible valuation.

What to watch: Watch organic growth and the prices paid for acquisitions.

Main risk: The European industrial cycle feeds through, the supply of suitable targets is finite, and Sweden withholds tax.

Leeway Rating

General scores - independent of the research topic

Leeway Score47.3/100

  • Business Rating 56.0
  • Market-Fit Rating Trend−1556.0
  • Cycle Rating 29.8

Check the full analysis

Constellation Software Inc.

CSU.TO · Technology · 66bn CAD

Long

Role in the portfolio: Constellation renews itself through spin-offs that arrive in the portfolio without a sale.

Why the stock fits: Constellation has already spun off Topicus and Lumine, whose shares land directly in the portfolio. The bet is that vertical software embedded deep in customers' workflows survives AI.

When buying stops: The position gets no new money if AI destroys customer switching costs or returns fall to the cost of capital.

Position and what to watch

Position: The position is about 1% and is deliberately a test position.

What to watch: Watch returns on new acquisitions after Mark Leonard's departure and the pace of further spin-offs.

Main risk: Vertical software is the textbook case of AI disruption fears. The premium valuation can shrink for years.

Leeway Rating

General scores - independent of the research topic

Leeway Score61.7/100

  • Business Rating 46.0
  • Market-Fit Rating Trend+574.4
  • Cycle Rating 64.8

Check the full analysis

Partners Group Holding AG

PGHN.SW · Financial Services · 16bn CHF

Long

Role in the portfolio: Partners Group opens access to winners that are not yet listed today.

Why the stock fits: Partners Group earns fees and carried interest on value created before companies go public. That addresses the risk that tomorrow's winners grow privately before an index or an investor can buy them. The high payout feeds renewal.

When buying stops: The position gets no new money if returns and fundraising in private markets fade for good, or fee pressure accelerates.

Position and what to watch

Position: The position is about 1% and is bought at a sensible valuation.

What to watch: Watch fundraising, fee rates and the realised returns of the funds.

Main risk: Switzerland withholds 35% from foreign investors, and reclaims are slow. Valuations of private assets are hard to verify. That future winners will emerge privately is an assumption, not an established fact.

No data

Your knowledge edge

Understand the real drivers of our portfolio

Look behind the scenes: see not only what we buy, but why. Get the complete view of our strategy.

  • Every position revealed: Access each holding in our portfolio.
  • Understand the investment thesis: See the exact role each security plays in our return.
  • See the connections: Analyse our long and short strategies in direct context.

Four paths through four decades

The weights are working assumptions, not predictions. The fourth path depends not on markets but on the tax home.

PathWeightWhat happens
Favourable: AI gains stay with shareholders 22% Productivity gains from AI accrue to shareholders, and real rates fall without repression. Several holdings and later additions grow 30 to 100 times, and the groups manage their succession. The portfolio earns about 9 to 11% nominal a year and ends 0 to 2 percentage points ahead of the global index after tax. The US underweight can hold it back, however.
Base: discipline first, then leaky repression 48% US valuations fall towards 20 times earnings, and bottlenecks are valued like utilities. One or two big winners give back their lead. 25 to 35% of the companies are cashed out, at times chosen by others, and the money is reinvested by rule. The portfolio earns about 7 to 8.5% nominal a year, close to the global index and clearly ahead of active funds.
Unfavourable: no new money and a sovereign debt accident 20% Without contributions the portfolio freezes into a snapshot of 2026. Succession failures, compulsory investment and AI hit large holdings, a sovereign debt accident forces austerity, and valuations fall to 11 to 14 times. Frozen assets or capital controls cost 5 to 15% of value. The portfolio earns about 3 to 5% nominal a year and lags the index by 1 to 3 points.
Political extreme: a tax on unrealised gains 10% The home state taxes unrealised gains every year, or levies a wealth tax above the portfolio's cash yield. The ban on selling turns into scheduled forced selling, and the tax advantage disappears. What is left is a stock index with high tracking error and frictions. The outcome depends on the tax home, not on markets.

Counter-arguments and Risks

These arguments result from stress-testing our fundamental assumptions.

The strongest counter-argument comes from the other side of the table: leave exceptional businesses alone. Acquisition programmes and constant restructuring can destroy more capital than any technological change. A broad, unchanged basket from 2026 keeps its few extraordinary winners while the losers become irrelevant. Valuation discipline and renewal rules, by contrast, could starve exactly those winners and keep buying mediocrity.

The US underweight is also open to attack. The valuation premium of American stocks may be earned: through higher returns on equity, deeper capital markets and the fact that the profits from AI are captured there. In an AI-driven disinflation the underweight could cost 1 to 2 percentage points a year. More fundamentally: because funds are banned, any selection of 50 to 80 stocks carries tracking risk against the index that need not be positive. The fund ban itself is the costliest rule of the whole approach.

Unresolved Market Factors

Open questions that cannot be conclusively answered using currently available market data.

The open question concerns the character of the approach: does self-binding prevent more destructive decisions than it forbids necessary ones? That applies above all once a single position makes up a large part of the wealth. Steering the inflows eases the problem. It cannot solve it when inflows are small relative to the grown portfolio.

Three unknowns come on top. Will the state reclaim the earnings of bottlenecks and record margins gradually, through falling valuations, or abruptly, through special taxes and price caps? Is the sequence of discipline followed by repression right, or will AI-driven disinflation or austerity win out? The post-war era shows how long repression can last: for decades, real returns on government bonds stayed below zero in many countries.13 And finally: is the investor's own tax home one in which a portfolio without sales remains affordable at all? No historical statistic answers this. Only a backtest free of survivorship bias, with matched contributions, actual corporate actions and withholding and wealth taxes, could settle it.

What investors should watch

Five measures show early whether the arithmetic works. They steer new money, not sales.

  • The trailing twelve-month P/E and dividend yield of the MSCI indices for the US, Europe, Japan, India and emerging markets before each purchase tranche. Forward earnings do not count.
  • Capital spending by the large cloud providers against their depreciation, and cash after all investment at the bottleneck companies in the portfolio.
  • Risk premiums on long-dated government bonds: the spread between French and German bonds, ten-year Japanese yields above 3%, and US and UK yields around 5%. Yield caps or mandatory purchases by domestic investors would mark the turn to repression.
  • New laws taxing holdings: limits on the US step-up in basis at death, taxes on unrealised gains, the Dutch Box 3 reform, new wealth taxes in Europe and changes to UK inheritance tax.
  • Rate cases and special levies on railways, exchanges, industrial gases and grids. They show whether bottlenecks keep their high returns or are regulated like utilities.

Sources

The numbered references in the text link to these sources. Estimates and derived values are marked as such in the text.

  1. iShares MSCI ACWI ETF: portfolio characteristics, trailing P/E on 22 October 2026
  2. Barron’s Market Data: S&P 500 P/E on 16 October 2026
  3. Bessembinder et al.: Long-Term Shareholder Returns, Evidence from 64,000 Global Stocks
  4. PwC Worldwide Tax Summaries: withholding tax rates
  5. Country Tax Calculator: wealth tax countries 2026
  6. Country Tax Calculator: guide to the Dutch Box 3 reform
  7. AAII: Valuing Growth Stocks, Revisiting the Nifty Fifty
  8. AMC Entertainment: completion of its debt refinancing, October 2026
  9. Journal of Financial and Quantitative Analysis: Short Squeezes and Their Consequences
  10. ECGI: Loyalty Voting Structures
  11. Investor.gov (SEC): Investor Bulletin on the escheatment process
  12. NIESR: Term Premia Tracker, third quarter 2026
  13. BlackRock: Fiscal repression

How this analysis is produced

The assessment is produced in several steps. Independent model families answer the same question separately and then attack the results. What you read here has survived several rounds.

  1. Two independent first theses. The same opening question goes to several model families that cannot see one another. Disagreements are kept, not averaged away.
  2. Dated evidence. Every claim that depends on facts is broken into individual search questions and answered with dated, sourced web research. Question, answer, sources and timestamp are logged and remain traceable.
  3. Adversarial review. Several review roles attack the thesis from different angles: one hunts for the strongest refutation, one for the awkward edge cases, one tests whether a path from thesis to share price exists at all, one checks the timeline for contradictions. Each role raises its own questions, which are again answered with evidence.
  4. Merge, then the next round. The surviving theses are merged into one and attacked again. The counter-position and the unresolved tension on this page come out of that step. They were not bolted on afterwards to look balanced.
  5. Back to the start. The process runs again until there is a clear result and a list of tradable companies with structural advantages.

Any analysis can be wrong. That is why the falsification criteria and the counter-position sit on the same page as the thesis, not in the small print.

The numbers shown against individual companies do not come from this process. The Business-Rating scores business-model quality, the Market-Fit-Rating eighteen fundamental figures against the current market, the Cycle-Rating the valuation against the stock’s own history. They are documented under the Leeway scores.

Questions before the first purchase

Why not simply buy a global index fund and never sell it?

That would be cheaper and easier. A global index fund renews itself from within and does not deviate from the market. This analysis deliberately tests a portfolio of single stocks, and the fund ban is its costliest rule. It does bring advantages: in Germany single stocks trigger no Vorabpauschale, they earn loyalty bonuses and double voting rights, and they can be bought selectively outside the most expensive parts of the market. Investors who allow funds will find the plan for that in our analysis A portfolio for three generations.

What happens when one stock makes up a third of the portfolio?

It stays, because an index does not trim its winners either. New money then goes to other positions and to newcomers, so its weight declines over time. That only works as long as inflows are meaningful relative to the portfolio. Winners can give back their entire lead, as Nortel, General Electric and Japan's giants after 1989 showed. This concentration risk is the weakness the approach knowingly accepts.

Who is the approach not suitable for?

It does not suit investors whose tax home taxes unrealised gains every year or levies a wealth tax above the portfolio's cash yield. Nor does it suit those who need to withdraw large sums regularly, or who can neither make new contributions nor hold a reserve for taxes and rights issues. In those cases everyday life forces sales, and the portfolio loses its edge.

Check the companies with Leeway's three ratings

The Business Rating assesses the quality of the business model, the Market-Fit Rating tests the fundamentals in the current market regime, and the Cycle Rating places the valuation in the stock's own history. For a portfolio that keeps its entry price forever, that third number deserves a look before every purchase.

Company Valuation and Fundamental Analysis

The data is recalculated on a weekly basis and depends on the current market value of the company and the balance sheet figures of the annual financial statements. The market value changes continuously with price changes, the balance sheets are created annually and change the valuation massively. The time of the annual financial statements and the metrics used can be viewed under "Metrics". Further information on how the analyses work can be found as tooltips directly on the analyses as well as in our explanations.

General

The information is intended to support you in your independent decision-making in implementing investment strategies and serves purely informational purposes. Past performance is not a reliable indication of future performance. No warranty can be given for the completeness, correctness and accuracy of the listed content. The information does not constitute specific investment recommendations. We neither know you nor your financial situation and do not provide investment advice. Only licensed investment advisors with knowledge of your personal circumstances may do this. PWP Leeway UG (limited liability) is not an investment advisor and does not collect any personal data for the purpose of investment optimization. PWP Leeway UG (limited liability) is a provider of investment recommendations and investment strategy recommendations. As such, it is registered with and supervised by the Federal Financial Supervisory Authority (BaFin). The legal notices on the use of the website and the General Terms and Conditions of PWP Leeway UG (limited liability) also apply.

All price data are closing prices of the respective stock exchanges. Price information and master data are provided by an external service provider. Furthermore, public trading data, such as provided by Finra.org, is used to analyze market behavior. No warranty can be given for the completeness, correctness and accuracy of the listed content.

Disclosure of interests: To the best of our knowledge, Leeway and the persons responsible for creating or publishing the respective content do not hold a net long or net short position exceeding 0.5% of the issued share capital of any issuer discussed in that content at the time of publication. There are no other interests or business relationships that could conflict with objectivity, unless expressly stated otherwise in the respective content.

© Leeway
PWP Leeway UG (haftungsbeschränkt)
Leeway Icon