MercadoLibre: An Investment Analysis Through the Lens of a Long-Term Business Owner

Cost structure and dependencies. The company's main costs are not mysterious: logistics and carriers, warehousing, first-party product costs, payment acquiring fees, sales taxes, risk control and bad-debt provisions, hosting/cloud infrastructure, plus product, technology, and marketing spending. It does not depend on a few large customers, but it does depend on a few key countries and institutional environments—Brazil, Mexico, and Argentina are especially important to revenue. At the same time, payment licenses, user-fund segregation rules, credit ceilings, and local tax regimes all affect margins and capital efficiency. In other words, customer concentration is low, but geographic and regulatory concentration is not.

Is this a business I can understand. If you understand it as "Latin America's version of Amazon + PayPal/Square + part of a bank/consumer-credit business," the broad strokes are understandable; but if you want to judge precisely the quality of every dollar of free cash flow, you must understand customer float, loan-asset growth, bad-debt provisions, regulation-restricted funds, and currency effects. This makes it understandable but not simple. If the stock market were to close for the next five years, I would be willing to own the business itself; but I would not ignore today's purchase price. Business understandability score: 4/5.

Industry and competitive landscape. The industry is in a long-term growth phase, not a mature harvesting phase. As MercadoLibre's IR materials disclose, Latin American e-commerce penetration is still only a "mid-double-digit percentage" of total retail, and third parties forecast the market will grow from $151 billion to $232 billion between 2023 and 2028. On Latin American financial services, MercadoLibre also repeatedly stresses that account, credit-card, and consumer-credit penetration in Mexico and Argentina remains low. An IDB/Finnovista report shows that the number of fintechs in Latin America grew to 3,069 between 2017 and 2023, indicating both enormous demand and steadily intensifying competition.

Main competitors and industry position. The company's own 10-K defines its competitors clearly: in e-commerce it faces Amazon, Shopee, local retail platforms, and cross-border platforms; in payments/finance it faces traditional banks, wallets, acquirers, card networks, QR-code payments, and cash itself. Third-party market data shows that in Brazil, Mercado Livre still holds the top position; Trade.gov, citing local industry sources, puts its market share at about 35%, with Amazon Brazil and Shopee close behind. In Mexico, COFECE's investigation concluded that Amazon and Mercado Libre together control more than 85% of online retail transactions, indicating that the company is at least not a marginal player in its two core markets but a leading platform that "regulators must keep an eye on."

Assessment of industry attractiveness. This is "a good company in a good industry," not "an excellent company in a bad industry." But it is not a good industry free of disruption risk: large technology platforms do not face a high barrier to entry, and the 10-K even states plainly that "for large, established technology companies, the barriers to entering some of our businesses are relatively low." At the same time, regulators in multiple countries are continually refining requirements on payments, fund segregation, credit-card rates, algorithmic transparency, and logistics bundling. Industry attractiveness score: 4/5.

The Moat

MercadoLibre's moat is "multi-layered and stacked," not a single brand premium. A brand advantage exists, but it shows up less in single price increases the way Coca-Cola's does and more in being the "default entry point" and in user trust. Scale advantages and distribution advantages are real: the company operates 18 e-commerce markets and 8 fintech markets in Latin America, its logistics network already covers shipment of the vast majority of items sold, in 2025 the network absorbed roughly 41% of incremental parcel volume over the full year, and nearly 75% of express deliveries arrive within 48 hours. Scale brings lower cost per parcel, faster delivery, and higher merchant/buyer satisfaction.

Network effects are central, but they need to be stated precisely: these are not "single-sided, social-network-style" network effects but the compound network effects of a "transaction platform + payment tool + logistics + advertising + credit." The more buyers, the more sellers want to come; the more sellers, the higher the advertising inventory and logistics density; the higher the payment and credit penetration, the higher the purchase frequency and repeat purchases; recurring interaction generates more first-party data, which in turn improves risk control and advertising efficiency. The company discloses that its credit models rely on merchant sales and consumer behavior data on the platform, something traditional banks inherently lack.

Data advantages and operating capability may be the hardest parts to replicate. MercadoLibre does not win on patent barriers but on local execution, risk models, fulfillment systems, and the ability to adapt to multi-country regulation. One important fact the 10-K reveals: regulatory licenses, user-fund segregation, capital and reserve requirements, KYC/AML, and the differing payment regimes across countries all require continuous investment; and operationally, the company keeps its technology architecture, development teams, and local product capabilities largely in-house. This barrier is deep against "new startups" but not as deep against "Amazon or large banks."

Switching costs are moderate to strong. It is not impossible for consumers to switch platforms, but if a merchant has already plugged payments, advertising, warehousing, fulfillment, credit, and traffic into MercadoLibre/Mercado Pago, the friction of leaving is significantly higher than being listed on a single platform alone. The more mature advertising and credit become, the stronger this multi-product binding. In 2025 and early 2026 the company emphasized record NPS in multiple countries and pointed to acquiring, software, and credit as important levers to reduce merchant churn, indicating that its moat is not merely "traffic" but "depth of ecosystem usage."

Is the moat widening, stable, or narrowing. My judgment is: still widening, but at the margin it now requires more continuous investment to maintain than it did two or three years ago. Advertising, credit, logistics, and the wallet add thickness to the flywheel; but Brazil's price war, cross-border low-price platforms, Mexico's antitrust scrutiny, and payment regulation all remind you that this is not a moat that automatically widens. For a competitor to replicate "a Latin American local ecosystem with equivalent stickiness" would take years, enormous capital, licenses, and local execution; but replicating just one link—say low-price traffic, acquiring hardware, or some seller subsidies—does not require the same price.

Pricing power, inflation, and recession behavior. It has indirect pricing power, not the classic direct pricing power of a luxury brand. The company can raise its monetization rate per dollar of GMV by deepening advertising penetration, payment penetration, logistics service depth, and credit penetration; but in reality it often reinvests part of that capability into lower free-shipping thresholds and higher subsidies to win share. In the first quarter of 2026, the company stated plainly that its operating margin fell from 12.9% to 6.9%, mainly because Brazil lowered its free-shipping threshold, shipping costs rose, and the expansion of the credit-card mix increased bad-debt provisions. This shows it can still be profitable during downturns or intensifying competition, but margins are not set in stone. Moat strength score: 4/5.

Management and Capital Allocation

The overall impression management gives is: long-term oriented, aggressive, but not "pursuing scale at any cost." Founder Marcos Galperin transitioned from CEO to Executive Chairman on January 1, 2026, with Ariel Szarfsztejn—cultivated internally over many years—taking over as CEO, an "orderly handoff" rather than a passive replacement. The 2026 proxy statement also shows that the Galperin-related entity, the Galperin Trust, reported holding about 7.0% of shares, indicating the founder's economic interests remain highly aligned with long-term shareholders, only that the stake is held mainly through a trust/entity rather than directly by the individual.

On capital allocation, over the past few years the company has reinvested the overwhelming majority of its cash into the business rather than paying dividends or making large buybacks. In its 10-K the company states explicitly that it suspended dividends starting in the first quarter of 2018, because management believes reinvesting capital into the business creates higher shareholder returns; buyback amounts over the past three years have also been modest—about $356 million in 2023, and nearly negligible in 2024 and 2025. This shows that management does not prop up EPS through financial engineering but prefers to put money into logistics, technology, credit, and international expansion.

On the question of "whether capital is allocated rationally," I give a positive assessment, but not a perfect score. The positives: acquisitions are very restrained—acquisition cash outflows were only $6 million in 2024, with no obvious large acquisitions in 2025; the bulk of capital expenditure goes into logistics and technology rather than flashy consolidations. The cautions: credit businesses naturally make "reinvestment returns look very high," but once risk control loosens or the regulatory environment shifts, returns on capital can reverse quickly, so management's capital-allocation ability must continually be tested by bad debt and funding costs.

Another reason my trust in management is fairly high is that in its shareholder letters it proactively explains the non-obvious factors behind profit volatility. For example, the company repeatedly explains that the credit-card business will first depress margins during its rapid-expansion phase because of upfront provisioning, but that a mature customer base reaches NIMAL breakeven within 12 to 18 months; this does not mean the risk disappears, but it shows management is willing to lay out the economics of the business clearly rather than only emphasizing GMV and revenue growth. Management and capital allocation score: 4/5.

Financial Quality

The table below is organized on the company's disclosed basis, in billions of dollars except for percentages; 2021–2023 data comes mainly from the 2023 10-K, 2024–2025 from the Q4 shareholder letter/annual report disclosures, and the latest quarter from the Q1 2026 report. Note: MercadoLibre's operating cash flow and traditional FCF are significantly distorted by customer funds, loan-asset growth, and related financing, so I also present "management-adjusted free cash flow."

This table reveals three key facts. First, growth is very strong: the 2021–2025 revenue CAGR is about 42%. Second, margins are not on a one-way climb: gross margin has fallen back from its 2023 peak, and the operating margin in the first quarter of 2026 compressed noticeably, showing that growth is not free. Third, accounting cash flow is strong but cannot be taken directly as "cash distributable to shareholders": 2025 operating cash flow reached $12.116 billion, yet "adjusted free cash flow" was only $1.481 billion, because float restrictions, loan growth, and related fund deployment consumed a large amount of the headline cash.

Are the profits real. I would rather put it this way: the profits are real, but you have to use the right basis. MercadoLibre's net income is not "padded profit," because in recent years net income, operating income, and shareholders' equity have all improved in step, and acquisitions and buybacks have not been used to dress up per-share figures; but if you simply take GAAP operating cash flow minus capital expenditure, you will severely overstate distributable cash, because the fintech subsidiary's customer funds, regulation-restricted investments, and loan-asset expansion are all baked in. The fact that management separately discloses adjusted free cash flow is precisely a sign that it knows the traditional FCF metric would mislead investors.

Return on capital. Traditional ROE is very high: based on a rough estimate using year-end equity, ROE in both 2024 and 2025 was at a very high level; Nu Holdings officially reported FY2025 ROE of 33%, and MELI, driven by its integrated platform and credit business, also exhibits high-return characteristics. It must be stressed that MercadoLibre's group-level ROIC/ROE is affected by customer funds, loan assets, and foreign-currency translation, and cannot be read mechanically as you would for a pure manufacturer; my approximate calculation shows that operating ROIC based on 2025–2026 TTM is still broadly in the high-double-digit range, but a precise point estimate should not be held with too much confidence.

Balance sheet and debt-servicing capacity. On the surface, the debt-to-asset ratio is very high, because customer funds, payables on card transactions, and related restricted cash are all on the balance sheet; but for operating safety, the more important metrics are net debt, Adjusted EBITDA, and interest coverage. In 2024 the company said its leverage ratio was below 1x and received an investment-grade upgrade from Fitch; by the first quarter of 2026, the company-defined net debt was $5.748 billion. On my rough calculation of TTM Adjusted EBITDA on the official basis, net debt/EBITDA is about 1.5x and interest coverage about 16x, indicating that this is not a fragile, highly leveraged story.

Working capital, inventory, receivables, and accounting risk. Inventory is small; what really deserves attention is credit-related assets. From 2023 to 2025, the company's net loans grew from about $2.694 billion to about $9.365 billion; in 2025 the credit portfolio expanded to $12.5 billion, up 90% from the prior year, which explains why bad-debt provisions and funding needs are always a core variable in valuation. My conclusion is: I currently see no typical financial-fraud red flags, but there are two complications that must be acknowledged—first, during 2024/2025 there were reclassifications and comparability adjustments in how revenue and expenses are presented; second, CECL/upfront provisioning makes margins look worse during periods of rapid expansion and makes it easier for outsiders to misjudge.

Owner Earnings and Intrinsic Value

The facts. As of around the U.S. trading session on May 21, 2026, MELI's share price was about $1,677.26, with a market cap of about $85.06 billion.

Facts and conservative assumptions. MercadoLibre does not disclose a split between "maintenance capital expenditure" and "growth capital expenditure," so in the Buffett-style Owner Earnings estimate I take a conservative approach: treating all capital expenditure as maintenance capital expenditure. On a rough TTM basis, net income is about $1.92 billion, adding back depreciation and amortization of about $892 million, and subtracting total capital expenditure of about $1.358 billion, gives conservative owner earnings of about $1.45 billion. This basis converges roughly with the company's disclosed TTM adjusted free cash flow of about $1.367 billion, so I treat $1.4–1.5 billion as the more conservative range for the current true earning power. On this calculation, the current share price corresponds to more than 58x Owner Earnings; using management's adjusted FCF, it looks more like about 62x. For even the finest company, that is not a generous valuation.

Method 1: Owner Earnings discounting. The valuation below is not a "share-price forecast" but takes current conservative owner earnings as the starting point and models the compounding ability over the next ten years under three scenarios. The growth rate, discount rate, and terminal growth rate here are all assumptions, not facts; the factual portion comes only from the revenue, profit, cash flow, net debt, and capital investment already disclosed above.

My interpretation. The current price of about $1,677 is already very close to the valuation range of my optimistic scenario. This means: if you buy today, you are effectively betting that the company can sustain Owner Earnings compounding in the high double digits over the next ten years while competition and regulation do not deteriorate meaningfully. Such an outcome can happen, but for a conservatively tilted investor this is not a margin of safety; this is a "high price prepaid for excellent execution."

Method 2: Relative valuation. Comparing the current market cap roughly against the latest public financials, MELI trades at about 44.3x TTM net income, 2.7x TTM revenue, 12.6x FY2025 book equity, and around 23x TTM EV/EBITDA. By comparison, Nu Holdings officially disclosed FY2025 net income of $2.9 billion and ROE of 33%, trading at about 21.7x FY2025 net income on its current market cap; Sea Limited officially disclosed FY2025 revenue of $22.9 billion and net income of $1.6 billion, trading at about 31.9x FY2025 net income and 2.2x FY2025 revenue on its current market cap; Amazon's current trailing P/E is about 38.6x. So MELI is not absurdly expensive, but it does depend more on future execution than most comparable names, and it is by no means "cheap because the whole peer group is expensive."

Method 3: Asset or liquidation value. This method has limited reference value for MELI, for two reasons. First, the customer funds within the payments business do not belong to shareholders—Brazilian regulation explicitly requires user funds to be segregated from the platform's own assets, with corresponding reserves maintained. Second, what truly holds value is the network, brand, data, fulfillment network, and user mindshare, not separately sellable land or inventory. By the end of 2025, shareholders' equity was about $6.748 billion, far below the current market cap; in other words, if what you are buying is "liquidation protection," MELI is not a good target. If what you are buying is "long-term compounding ability," then you must have very high confidence in growth and competition.

Margin-of-safety conclusion. My overall judgment is: the conservative intrinsic value range is about $400–600; the reasonable intrinsic value range is about $750–1,000; the optimistic intrinsic value range is about $1,350–1,850. At the current price, the stock carries no discount to "fair value" and is in fact closer to "optimistic-scenario pricing." The margin of safety is therefore insufficient. If you insist on a margin of safety of at least 20%–30%, I would treat $800–1,000 as the range to start studying seriously, $600–800 as the more attractive buy range, and above $1,600 I would lean toward defining as the "clearly not cheap" range.

Open questions and limitations. Three points must be stated frankly here: the company does not disclose maintenance capital expenditure, so Owner Earnings uses "total Capex in place of maintenance Capex," making the conclusion conservative; group-level ROIC is distorted by the accounting basis of customer funds and the loan business, so I trust the trend more than a point estimate; some peer metrics in the relative valuation use each company's latest public annual report/earnings disclosure, and the bases are not fully consistent. This uncertainty itself is one reason I am unwilling to issue a "Buy" at the current price.

Risks, Comparisons, and Final Judgment

The most important risks and the strongest bear case. The strongest bear logic is actually very clear: this may be a great company, but it may already be priced on an ideal script of "sustained high growth over the next ten years, controllable bad debt, defensible competition, and bearable regulation." If any one link breaks, valuation compression and earnings downgrades will occur at the same time. Specifically, competitive risk comes from Amazon, Shopee, Temu, and local platforms attacking on logistics and the low-price band; technology and business-model risk comes from advertising, payments, and credit each potentially being carved out in part by more specialized players; regulatory risk concentrates on payments, fund segregation, rate caps, algorithmic transparency, and logistics bundling; financial risk shows up mainly when the credit portfolio grows too fast, with provisions and funding needs causing margins to swing sharply.

Which facts would overturn the investment judgment. If the following facts emerge in the future, I would admit the original judgment was wrong: first, buyer/merchant frequency and NPS stop improving, indicating ecosystem stickiness is weakening; second, the credit portfolio keeps growing rapidly, but the 15–90-day delinquency rate and NIMAL deteriorate simultaneously, indicating a problem with growth quality; third, margins in Brazil and Mexico still fail to return to a reasonable range after subsidies normalize, indicating the price war has become structural; fourth, regulators force the platform to unbundle key linkages, weakening the synergy among payments, logistics, and traffic. In 2024 the company's total credit portfolio was about $6.6 billion with a 15–90-day NPL of about 7.4%; by the fourth quarter of 2025 the total credit portfolio grew to $12.5 billion and the credit-card portfolio's 15–90-day NPL fell to 4.4%, indicating that risk management holds up for now, but also that it must be watched continuously over the next two quarters to two years.

Comparison with the index, the risk-free rate, and other opportunities. Buying MELI at today's price, you have to compare it against simpler alternatives. The U.S. 10-year Treasury yield was about 4.57% on May 20, 2026, while my expected ten-year annualized return for MELI at the current price is roughly: about 3% in the conservative scenario, about 7%–8% in the neutral scenario, and about 12% in the optimistic scenario. This means that in the conservative scenario your return is even worse than long-term Treasuries; in the neutral scenario the excess return over the risk-free rate is not dramatic either, yet you take on clearly higher regional, currency, credit, and single-company risk. For a conservatively tilted investor, this does not constitute a buy point that is "clearly superior to the index."

Investment Checklist. Based on the above analysis, my judgments are as follows.

Final investment conclusion.

【Final Rating】 Watch

【One-Sentence Investment Thesis】 MercadoLibre looks like one of the most valuable pieces of digital commercial infrastructure in Latin America, but today's price looks more like buying "sustained excellent execution over the next ten years" than buying an asset that "leaves a sufficient margin of safety."

【Core Bull Case】

• E-commerce, payments, logistics, advertising, and credit form a strong flywheel, and the compound moat is deeper than that of a single platform.

• Revenue has compounded at an extremely high rate over the past five years, and the first quarter of 2026 still posted 49% year-over-year growth.

• The logistics network, payment penetration, and data-driven risk control keep raising the value per user.

• Management is clearly long-term oriented, with restrained acquisitions and disciplined reinvestment.

• The balance sheet remains sound, with net debt/EBITDA and interest coverage both manageable.

【Core Bear Case】

• The current valuation is already close to optimistic-scenario pricing, with an insufficient margin of safety.

• Margins are highly sensitive to Brazil's free-shipping policy, logistics investment, and credit provisions.

• The fintech and credit businesses make true distributable cash flow far below headline OCF.

• Latin American currency, tax, and regulatory disruptions are significant.

• While competition has not destroyed the moat, it is enough to prevent the company from easily "earning in its sleep."

【Key Assumptions】 For this investment to work, it must at least satisfy: the company can still sustain high-double-digit Owner Earnings growth over the next ten years; the credit-portfolio expansion does not trigger persistent runaway bad debt; competition in Brazil and Mexico does not keep the operating margin depressed at mid-single digits for the long run; regulation does not force a significant weakening of ecosystem synergy.

【Fair Buy Price】 The more attractive range to research/build a position is $800–1,000; if a systemic drawdown drives it down to $600–800, I would treat it as closer to a "buy point in the value-investing sense." The basis for this range is my required discount to the conservative and reasonable intrinsic value ranges, not a short-term technical level.

【Target Holding Period】 If you buy, do so on the premise of 10 years or more; companies like this realize their value most readily only through long-term compounding.

【Expected Annualized Return】 Estimated from the current price: about 3% in the conservative scenario, about 7%–8% in the neutral scenario, and about 12% in the optimistic scenario. For a conservatively tilted investor, this distribution of returns is not enough to support "buy now."

【Maximum Loss Risk】 If competition intensifies, bad debt rises, margins step down, and the valuation compresses from 44x earnings to a more normal range, I think a permanent share-price loss of 40%–60% over the coming years is not unimaginable; in an extreme case, if ecosystem synergy is broken, the loss could be higher. The source of this risk is not short-term volatility but a fundamental stall after "a high-quality company is bought at a high valuation."

【Tracking Metrics】 Going forward I will keep tracking: FX-neutral revenue growth; the operating margin and its recovery after Q1 2026; Mercado Pago MAU, active buyers, and advertising penetration; total credit-portfolio growth, 15–90-day NPL, and NIMAL; Adjusted free cash flow; net debt/EBITDA; Brazil's free-shipping and logistics costs; regulatory developments in Mexico and Brazil; changes in share count and whether aggressive acquisitions appear.

【Signals That Trigger a Reassessment】 If any of the following occur, I will immediately review again: the credit portfolio keeps expanding rapidly while bad debt and provisions deteriorate in tandem; margins stay at an overly low level for several consecutive quarters; user/merchant frequency or NPS visibly weakens; regulation weakens the linkage among payments, logistics, and traffic; management begins to explain growth through large-scale acquisitions or aggressive accounting.

【Final Recommendation】 Put plainly, MELI is worth keeping on a high-priority watch list for the long term, and even worth spending more time to understand than the average company; but if you insist on "the perspective of a business owner with a sufficient margin of safety right now," I do not recommend rushing to buy at the current price. This is an excellent business, not an excellent entry point.

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.