# Nomad Partnership _Scale economics shared is the program that keeps on giving._ --- ## Letters nobody read Between 2001 and 2014, Nick Sleep and Qais Zakaria wrote biannual letters to the investors in their Nomad Partnership. The letters circulated quietly among a small group of sophisticated allocators. Almost nobody else noticed. The fund returned 921% over thirteen years - versus 117% for the MSCI World Index. By the end, the portfolio held just four stocks: Amazon, Costco, Berkshire Hathaway, and Liberty Global. Sleep and Zakaria then did something unusual: they closed the fund, returned nearly $8 billion to investors, and walked away to focus on philanthropy. The letters have since achieved cult status. John Collison of Stripe wrote the foreword to the published collection. The ideas have been dissected by investors worldwide. But most coverage focuses on Sleep as a stock picker. The more interesting question is whether his core insight - scale economics shared - works as an operating strategy, not just an investment thesis. And whether more companies could adopt it if they chose to. --- ## Scale economics shared Sleep coined the term "scale economics shared" in the 2008 letter: "As the firm grows in size, scale savings are given back to the customer in the form of lower prices. The customer then reciprocates by purchasing more goods, which provides greater scale for the retailer who passes on the new savings as well." This creates a flywheel. Scale reduces unit costs, lower costs enable lower prices, lower prices attract more customers, more customers create more scale. The cycle repeats. What makes this distinctive is the choice about where value accumulates. Most companies treat [[Scale]] economies as profit to be captured - margins expand as the business grows. Sleep's insight is that sharing the gains with customers can be more valuable than keeping them. The customer reciprocation compounds. "This is why firms such as Costco enjoy sales per foot of retailing space four times greater than run-of-the-mill supermarkets," Sleep wrote. "Scale economics shared incentivises customer reciprocation, and customer reciprocation is a super-factor in business performance." The companies that do this - Costco, Amazon, Walmart, GEICO - don't just grow. They become destination businesses that customers actively seek out. The competitive moat isn't just scale; it's the behavioural loyalty that scale economics shared creates. --- ## Ten years later Sleep closed the fund a decade ago. The thesis has had time to be tested. **Costco continues to execute perfectly.** The $1.50 hot dog and soda has been the same price since 1985. Gross margins remain capped below 12% - not because Costco can't charge more, but because it won't. When inflation hit multi-decade highs in 2022, Costco's CFO rattled off price reductions: batteries down $2, reading glasses down $2, sporting goods and lawn care set lower year-over-year. "We always want to be the first out there trying to lower prices," he said. The company even operated its own shipping fleet during the pandemic to avoid passing supply chain costs to members. It's institutional, not personal - a textbook example of [[Process power]]. Jim Sinegal retired as CEO in 2012. The strategy hasn't wavered. **Amazon has cut AWS prices over 100 times since 2006.** The cloud business could capture more margin - it's the profit engine funding everything else. Instead, Amazon shares the efficiency gains: 45% reduction on GPU instances in 2025, 85% reduction on S3 Express storage. "Your margin is my opportunity," Bezos told suppliers. The same logic applies internally. AWS margins fund customer acquisition elsewhere. Prime members saved $95 billion through free delivery in 2024. Same-day delivery sites expanded 60%. The flywheel keeps spinning. **The returns validate the thesis.** An investor who held Costco, Amazon, and Berkshire Hathaway from 2014 would have dramatically outperformed almost any alternative portfolio. The concentrated bet on scale economics shared worked precisely as Sleep predicted. --- ## New adopters The model has spread to new markets and new companies - though adoption remains surprisingly rare. **Mercado Libre** is building in Latin America what Amazon built in North America fifteen years earlier. Unit shipping costs fell 8% quarter-over-quarter in late 2025 as scale kicked in. Three-quarters of orders now deliver within 48 hours, almost exclusively through their own logistics network. The efficiency gains fund subsidised shipping that competitors can't match. E-commerce penetration in Latin America remains below 16% - far behind mature markets. Mercado Libre is betting that sharing scale economics now will lock in the market as it develops. The playbook is familiar. **Coupang** in Korea followed a similar path. The company lost over $600 million annually building logistics infrastructure. Within three years, two out of five Koreans were customers. The losses were an investment in scale - a classic case of [[Paying for growth]] before capturing it. The scale now generates profits. Coupang is demonstrating that the model works outside the American context. **ALDI and Lidl** represent a different variant - scale economics shared through radical simplicity rather than technological investment, as [[Retail Disruptors]] documents. Ninety percent private label. No advertising in Germany. Cardboard displays. Customers bag their own groceries. Every saved cost becomes a lower price. The model built dominance in Europe and is now expanding aggressively into the United States. --- ## Temu: efficiency or subsidy? Pinduoduo's Temu is testing how far the model can be pushed - or perhaps how far it can be distorted. Temu is losing an estimated $8-9 billion annually to offer prices 40-60% below Amazon. Within two years, they onboarded 200,000 retailers and attracted 467 million users worldwide. The growth is extraordinary. But this isn't quite scale economics shared. Temu is subsidising prices with investor capital, not sharing genuine efficiency gains. The bet is that behaviour locked in now will generate margins later. "Lock in behaviour first, negotiate margins later" is the explicit strategy. Classic scale economics shared requires actual scale economies - real cost advantages from volume, logistics, purchasing power. Temu is buying market share with cash, hoping scale economies will eventually materialise to justify the investment. Whether this works depends on whether the efficiency gains ever arrive, and whether customers stay when subsidies end. The 2025 tariff changes pose an existential test: Temu's model depends on de minimis exemptions that may disappear. Scale economics shared through genuine efficiency is more durable than scale economics subsidised by investors. --- ## Why this remains rare The striking thing about scale economics shared is how few companies do it. The logic is clear. The evidence is compelling. Yet most businesses capture scale economies as profit rather than sharing them with customers. Structural barriers explain this. **Quarterly earnings pressure.** A McKinsey survey found that 55% of CFOs would pass up an attractive long-term investment to avoid missing quarterly earnings targets by even a small amount. Sharing efficiency gains with customers means lower reported margins. Lower margins disappoint analysts. Disappointed analysts move stocks. Executives whose compensation depends on stock price learn to prioritise margins over customer value. Long-term oriented companies create more value despite this pressure. McKinsey's research found they generated $7 billion more in market capitalisation between 2001 and 2014. But the path there requires surviving short-term pressure that most management teams and boards won't tolerate. **Incentive misalignment.** Executive compensation typically rewards earnings per share growth over three to five year windows. Scale economics shared sacrifices near-term earnings for long-term compounding. The strategy makes sense on a ten-year horizon; it looks like underperformance on a three-year horizon. Executives rationally optimise for the metrics they're measured on. **Ownership structure.** [[Semper Vic|Tom Russo]], who has run a concentrated portfolio of compounders for decades, notes that roughly 60% of his holdings are family-controlled businesses. The reason: families have the "capacity to suffer" through investment periods that public market shareholders won't tolerate. Credit Suisse research found that family-owned firms outperformed the market by about 3% annually between 2006 and 2022. The advantage is the freedom to think in decades rather than quarters, not operational brilliance. Costco's founding CEO Jim Sinegal famously told analysts who wanted higher margins: "This is not the business for you." Most CEOs can't say that. Their boards won't let them. **Capital market structure.** Activist investors demand margin expansion, not margin sacrifice. Passing savings to customers looks like "leaving money on the table" - a form of [[Counter-positioning]] that incumbents find structurally impossible to copy. The vocabulary of shareholder value has been captured by short-termism dressed as discipline. The companies that practice scale economics shared either have protective ownership structures (family control, dual-class shares), patient capital bases (Berkshire Hathaway as a shareholder), or founders with enough control to ignore the pressure (Bezos at Amazon for decades). **Requires scale you don't yet have.** You need economies before you can share them. The strategy requires surviving a period of lower margins while building the scale that eventually justifies them. Patient capital is required - and patient capital is scarce. --- ## Ownership as enabler Sleep's letters don't emphasise ownership structure explicitly, but the pattern is clear in his portfolio. Costco had Sinegal's founding vision embedded in culture. Amazon had Bezos with voting control. Berkshire had Buffett. Liberty had Malone. These aren't accidents. Scale economics shared requires someone with the authority to resist pressure for short-term margin capture. Family founders, dual-class structures, and concentrated ownership provide that authority. Dispersed public ownership rarely does. Practicing scale economics shared probably requires ownership or governance structures that protect against quarterly pressure. This is a question of [[Designing the organisation]] around the strategy, not hoping culture will carry it. [[TCI|Chris Hohn]]'s observation that management matters less for businesses with hard assets inverts here - for scale economics shared, governance structure matters enormously because the strategy requires resisting the default behaviour of public markets. --- ## Future evolution Several forces are reshaping the model. **Digital intensifies the economics.** Software and cloud services have extreme scale economies - marginal cost approaches zero. AWS can cut prices 100 times because each price cut barely affects unit economics while dramatically affecting competitive position. The flywheel spins faster in bits than atoms. AI and automation may accelerate this further. Every efficiency gain from machine learning is a potential price reduction. Companies that share these gains will compound customer loyalty; companies that capture them as margin will face competitors who don't. **Subscription models are a variant.** Netflix and Spotify don't cut prices with scale - they invest efficiency gains in content and catalog. The value shared with customers is better product rather than lower price. Same underlying logic: compound customer value rather than capture it as profit. **Threats to the model.** Deglobalisation and tariffs challenge companies built on global scale. Temu's entire model depends on supply chain configurations that trade policy may eliminate. Diseconomies of scale emerge at extreme size - bureaucracy, coordination costs, the weight of complexity. And the model requires discipline that can erode. When competitive pressure eases, the temptation to capture margins grows. Walmart's evolution from Walton's original vision illustrates how scale economics shared can fade into conventional margin maximisation. --- ## Conditions for adoption Sleep's insight started as an investment thesis: find companies practicing scale economics shared and hold them forever. But the deeper question is whether more companies could adopt the strategy - and what would have to change to enable it. The barriers are structural, not mysterious. Governance favours short-term thinking. Incentive systems reward margin over customer value. Capital markets punish investment periods. These are design choices, not laws of nature. Several conditions need to hold for the strategy to work. First, actual scale economies must exist. The strategy requires cost advantages that grow with volume. Service businesses with linear cost structures can't share economies they don't have. Second, patient capital must fund the investment period. Sharing gains with customers means lower near-term margins. The flywheel needs time to become self-sustaining. Third, governance must protect long-term thinking. Family control, dual-class shares, concentrated ownership, or a board explicitly committed to long-term strategy. Without structural protection, short-term pressure erodes the discipline. Fourth, incentives must align. Executive compensation tied to ten-year outcomes rather than three-year EPS. Metrics that track customer value, not just margin capture. Fifth, the discipline must institutionalise. Sinegal retired from Costco over a decade ago. The strategy persists because it's embedded in culture and process, not dependent on a single leader. Scale economics shared needs to survive succession. Most companies fail several of these conditions. That's why the strategy remains rare despite its obvious logic - the insight is simple, but the execution requires structural conditions that most organisations lack. The companies that do practice it - Costco, Amazon, Mercado Libre, ALDI - aren't just operationally excellent. They're structurally configured to resist the pressure that makes everyone else capture margins instead of sharing them. That structural configuration is the actual insight. Scale economics shared is the strategy. But the capacity to execute it is a governance and ownership design problem. ---