MercadoLibre (MELI)
E-commerce / Fintech (LatAm)
Latin America's leading e-commerce and fintech company, growing 46% in dollars (driven by Brazil and Mexico); deliberate investment pushed the operating margin down to 6.7% and operating income is falling even as revenue soars — the question is not growth, it is whether the margin comes back.
Moat Compounder estimates the intrinsic value of MercadoLibre (MELI) at $3,922 per share on a five-year horizon. With the stock at $1,892.44 at 2026-10-12 close, the expected total return is 15.7% per year: very undervalued. The analysis draws on 10-K FY2025 and 10-Q Q2 2026. Analysis dated 2026-08-06.
- Price
- $1,892.44
- Intrinsic value (5y, base)
- $3,922
- Total annual return (5y)
- 15.7%
- Status (nominal)
- Very undervalued
- Margin of safety
- +40%
The essentials
- Revenue +46% over the last twelve months; the second quarter grew 50%, the fastest pace in four years, though 43% on a constant-currency basis.
- The flip side: operating income fell 17% in the quarter and the margin came in at 6.7%. Over the last twelve months operating income is lower than a year ago.
- The incremental return on capital is negative (−1.1%): US$3.8bn of invested capital went in and profit did not rise. That is where the thesis is decided, not on growth.
Intrinsic value — two valuation methods
By both methods, the value today (DCF $2,316 · Multiples $3,147) exceeds the market price ($1,892).
Pillars of the analysis
The verdict — today vs 5 years
Today — with margin of safety: at $1,892 trades ~39.9% below its value discounted to today (~$3,147) — the wide discount we require (≥38%, equivalent to a ~15% annual return); the risk is covered by the margin, not the rate.
At 5 years — Muy infravalorado: the target price ($3,922) plus dividends yield above the required average return (10%) — the business compounds.
The bridge: the return at 5 years comfortably exceeds the risk-free rate (4.5%) — and the discount reaches the required margin of safety.
Thesis
The business
MercadoLibre is Latin America's leading e-commerce and fintech company, and it grew 46% in dollars over the last twelve months; the second quarter of 2026 was the fastest in four years. Growth is driven by Brazil and Mexico, which together contribute ~77% of revenue. At constant currency the consolidated figure grows 43%, so the growth is genuine and not an effect of foreign exchange or of Argentine inflation. It is a quality business, with a wide moat (network, in-house logistics, credit data and brand) and negative working capital that funds it on its own.
The valuation
The valuation is a sum of the parts: each line with its own method and multiple —commerce and advertising at EV/NOPAT (20×, a capital-light marketplace), Mercado Pago at EV/EBIT (17×, acquiring) and Mercado Crédito at a lender's P/E (15× normalized earnings, not a marketplace premium)—. The correct measure of cash is not reported operating cash flow —inflated by the float— but the adjusted free cash flow the company publishes, US$1.13bn over the last twelve months; that is why today's multiple (50× P/E) is misleading: it is the wrong lens for a business compressed by investment.
The projection assumes a gradual deceleration (~40% to ~20%, starting from the constant-currency rate rather than the headline) and a slow, partial margin recovery: next year it drops to 7.5% —the current run rate, not the last-twelve-month average— and only by the fifth year reaches ~11.5%, below the 13.5% assumed in the previous version and far from the historical peak of 14.6%. At ~$1,892 the sum of the parts anchors at ~$3,922 per share over five years, equivalent to an annual return (CAGR) of +16%. The discounted cash flow, brought to today at the method's risk-free rate, yields a per-share value above the market price and reinforces the reading from multiples.
The margin of safety
At ~$1,892, the expected return is +16%, and the margin against the 15% threshold demanded of a great investment is +3%. The verdict is Very undervalued. The gap between market perception —which extrapolates the margin decline and reads a voluntary investment as deterioration— and the reality of the business —whose growth is accelerating— is the source of the opportunity. The honest objection is that, for now, only the cost is visible: the incremental return on invested capital is negative.
What to watch
The real uncertainty is not growth, which is solid, but the margin recovery: the adverse scenario, with the margin stabilizing at ~8.5%, yields a return of -6%. It is the most sensitive assumption in the case, and the concrete test to follow is whether the quarterly margin stops falling: it was 6.9% and then 6.7%, with management describing it as broadly stable versus the prior quarter. The second front is the credit cycle: the book is growing 75% —29 points faster than revenue— with net interest margin after losses compressing to 20.7% and the card already in negative territory (−2.5%) because of the step-up in issuance. And one methodological caveat: because MercadoLibre reports by geography rather than by business line, the allocation of segment margins in the sum of the parts is an analyst estimate —the largest source of uncertainty in the case—.
Educational / informational. Does not constitute investment advice.
Valuation by multiples — sum of the parts
Discounted cash flow to present value (DCF)
Normalized owner earnings (margin partially recovered) as the base. Move the assumptions: the value recalculates live. The verdict remains anchored by multiples; the DCF contrasts it at present value.
Risk does not inflate the rate: protection is required separately, as a margin of safety over the value. The floor avoids discounting at the pace of a depressed market rate.
| Year | Projected FCF | Discount factor | Present value |
|---|---|---|---|
| 1 | $2.8 bn | 0.957 | $2.7 bn |
| 2 | $3.3 bn | 0.916 | $3.1 bn |
| 3 | $3.9 bn | 0.876 | $3.5 bn |
| 4 | $4.7 bn | 0.839 | $3.9 bn |
| 5 | $5.5 bn | 0.802 | $4.4 bn |
Reverse DCF — what growth the price discounts
The inverse approach: instead of projecting growth to obtain the value, the market price ($1,892) is taken as given and it solves for what annual owner-earnings growth would need to hold for 5 years for the present value —at the method's rate (4.5%, no-growth terminal)— to equal that price. It is the disconfirmation test: the expectations the price already pays for, contrasted against the method's projection.
The market discounts less growth (13.0%/year) than we project (18.0%/year) → if the base case holds, there is margin: perception is more pessimistic than the estimated reality.
That growth implies ~$4.4 bn of owner earnings in year 5 (vs ~$5.5 bn of our base case). It recalculates if the DCF assumptions are edited.
Year-by-year model — sum of the parts by segment
Each segment is projected and valued with its own metric and its own multiple (Commerce EV/NOPAT · Payments EV/EBIT · Credit lender P/E), and they are added up by year → the value curve. The multiple compresses toward the terminal (at 3 years there is more growth still ahead), so the terminal reproduces the SOTP. In edit mode, the metrics and multiples of each part can be adjusted.
| US$ bn | TTM | +1a | +2a | +3a | +4a | +5a |
|---|---|---|---|---|---|---|
| Consolidated operation (the top line that feeds the segments) | ||||||
| Revenue | 35.2 | 49.3 | 66.2 | 85.8 | 107 | 128 |
| growth | — | +40% | +34% | +30% | +24% | +20% |
| Operating margin | 8% | 8% | 9% | 10% | 11% | 12% |
| Capex | 1.5 | 2.0 | 2.6 | 3.2 | 3.9 | 4.6 |
| Commerce + Ads · EV/NOPAT · multiple 3 years 22.0x → 5 years 20.0x | ||||||
| NOPAT | 1.36 | 1.64 | 2.39 | 3.25 | 4.19 | 5.07 |
| multiple applied | — | 22.0x | 22.0x | 22.0x | 21.0x | 20.0x |
| EV = metric × multiple | — | 36.1 | 52.6 | 71.5 | 88.0 | 101 |
| Mercado Pago · EV/EBIT · multiple 3 years 18.0x → 5 years 17.0x | ||||||
| EBIT | 1.12 | 1.29 | 1.79 | 2.31 | 2.83 | 3.25 |
| multiple applied | — | 18.0x | 18.0x | 18.0x | 17.5x | 17.0x |
| EV = metric × multiple | — | 23.2 | 32.2 | 41.6 | 49.5 | 55.3 |
| Mercado Crédito · P/E norm. · multiple 3 years 16.0x → 5 years 15.0x | ||||||
| Utilidad norm. | 0.66 | 0.81 | 1.2 | 1.67 | 2.18 | 2.68 |
| multiple applied | — | 16.0x | 16.0x | 16.0x | 15.5x | 15.0x |
| EV = metric × multiple | — | 13.0 | 19.2 | 26.7 | 33.8 | 40.2 |
| Consolidated by year (sum of the parts) | ||||||
| Σ EV parts | — | 72.3 | 104 | 140 | 171 | 197 |
| + Cash − Debt = Equity | — | 74.3 | 106 | 142 | 173 | 199 |
| Value / share | — | $1,465 | $2,091 | $2,797 | $3,418 | $3,922 |
| CAGR vs price | — | (-23%) | (+5%) | (+14%) | (+16%) | (+16%) |
With today's market value fixed and the consolidated metric growing, the multiple the entry point implies each year. It is not the valuation (that is the sum of the parts above); it measures how expensive or cheap the entry point is: today's high multiple is the price of growth and it cheapens on its own.
| Consolidated EV / metric | TTM | +1a | +2a | +3a | +4a | +5a |
|---|---|---|---|---|---|---|
| EV / NOPAT | 46.0x | 36.3x | 23.6x | 16.3x | 11.9x | 9.1x |
| EV / EBIT | 32.2x | 25.4x | 16.5x | 11.4x | 8.3x | 6.4x |
| EV / owner-FCF (maint.) | 40.0x | 32.5x | 20.2x | 13.5x | 9.6x | 7.3x |
| EV / Sales | 2.7x | 1.9x | 1.4x | 1.1x | 0.9x | 0.7x |
Sum of the parts by segment, year by year: each line with its own metric and its own multiple (Commerce EV/NOPAT · Payments EV/EBIT · Credit lender P/E, on earnings with the expected loss provision normalized), added up by year → the value curve. The multiple compresses toward the terminal (at 3 years there is more growth still ahead). Cash does not accumulate: Payments and Credit redeploy the flow (float / loan book), and Commerce is capital-light. Base year to Jun-2026: consolidated trailing-twelve-month operating income ($2.91bn) is split keeping the proportion declared in the previous analysis, with two corrections the quarter does support. Commerce absorbs the compression —discounts for paying with PIX, the cut in seller fees in Brazil and higher shipping costs are the causes the company names— and falls to $1.36bn of NOPAT from $1.64bn. Credit rises to $0.66bn: the loan book passed $16bn (+75% year over year) and the company describes it as highly profitable, although the net interest margin after losses compressed 2 points, to 20.7%, on the shift toward the card (from 43% to 47% of the book), whose margin went from breakeven to −2.5% on the jump in issuance. Payments stays roughly flat at $1.12bn. ⚠️ MercadoLibre reports by geography, not by segment → the split by line is an analyst estimate (the largest uncertainty in the case).
Scenarios (bear / base / bull) — at 5 years
Value sensitivity
Value per share by growth scenario (rows) and the compression or expansion of the exit multiple (columns). The color shows whether it beats the required return.
| Growth ↓ / Multiple → | Compression−15% | Base multiple | Expansion+15% |
|---|---|---|---|
| BearRevenue +28→10% · base Multiples by piece · blended ~14.3× NOPAT | $1,205 -8.6% | $1,417 -5.6% | $1,630 -2.9% |
| BaseRevenue ~40→20% (FX-neutral) · base Multiples by piece · blended ~19.1× NOPAT | $3,334 12.0% | $3,922 15.7% · base case | $4,510 19.0% |
| BullRevenue ~44→25% · base Multiples by piece · blended ~21.9× NOPAT | $5,631 24.4% | $6,625 28.5% | $7,619 32.1% |
Multiples — today
High today = growth is being paid for; they cheapen toward 3 and 5 years (see Projections).
Forward multiples
With today's price fixed and the metric growing, what multiple is being paid at 3 and 5 years. Today's high multiple is the price of growth: if the business grows, the entry multiple cheapens on its own.
Optionalities
They are valued separately, with their own rationale, and are not incorporated into the base or the verdict (they are excess return). When assigning them value — in Editmode —, the total with optionalities updates live, without moving the base.
The verdict, the base CAGR, and the margin of safety are always calculated on the base; optionalities do not alter them (with optionalities at $0 they do not move).
Maximum price to pay today — by required return
Each card fixes a required annual return and answers: if the business is worth $3,922 in 5 years, what is the maximum that can be paid today to obtain that return? Since it now trades at $1,892, the margin of safety is how much cheaper the market is than that maximum. The three thresholds: 4% covers inflation (the floor), 10% is the long-term average return, and 15% is the level of a great investment.
Return and margin of safety calculator
The maximum price to pay today to earn the required return, with the dividend collected as a separate flow. Both controls are editable.
With a target price of $3,922 in 5 years and a required return of 4.5% annually, the maximum to pay today is $3,147. Against the current market price ($1,892), the margin of safety is 39.9% (trades below the maximum → there is margin) and the total return at that price would be 15.7% annually.
Valuation quality
- Anchor: forward multiples by line. Each business valued with its own method and explicit exit multiple, brought to today at the method's risk-free rate — this drives the verdict. Caveat: MELI reports by geography, not by line → segment margin is an estimate (the biggest uncertainty).
- The DCF converges. The DCF (present value, at the method's risk-free rate, terminal with no growth) broadly converges with the multiples brought to today → a second method that reinforces, not a standalone number.
- Today's multiple is misleading, cheapens forward. 50× P/E on earnings depressed by investment: it is the wrong lens today, and if the margin recovers the forward multiple cheapens considerably (9× EV/NOPAT at five years).
- Credit multiple — the most demanding assumption. The Credit piece is valued at 15× normalized earnings (lender discipline: normalized P/E, not a marketplace's premium). At 20× it would add ~US$260/share; the more conservative multiple was chosen because a credit book rarely sustains 20× given leverage, the cycle, and CECL.
- Margin of safety at the end. At the market price the base CAGR is +16%, with a margin of +3% against the 15% hurdle. The margin of safety lives in the required hurdle, not baked into each assumption.
- Optionalities kept separate. Normalized credit, ads, and digital bank are valued separately (default $0) → they don't contaminate the base; they're free upside.
- Sensitivity — what decides it. Value is highly sensitive to the margin recovery: the adverse scenario (margin ~8.5%) yields -6% annually versus +16% in the base case. The uncertainty is the margin, not growth.
ROIC vs the 10% bar — the compounding engine
The quality bar — return bands
The return on capital is judged against absolute bands; the value-creation floor is the market's opportunity cost (~10%). A stock's volatility does not measure business risk.
ROIC 17% → excellent (15-20%). The bar is a measure of business quality, not the method's discount rate: value is discounted to today at the risk-free rate, and protection is required separately, as a margin of safety.
Owner earnings — the waterfall
It charges maintenance capex (which EBITDA does not deduct). The growth capex ($0.7 bn) is voluntary and is not charged to the base — it depresses FCF today, creates value tomorrow.
Cash & reinvestment
Grows ~46% with an intrinsic growth of ~6%: the rest is financed by the float / capital-light structure (a cost-free funding advantage).
Margins — trajectory
Each margin over sales, year by year: historical (solid line) → projection (dotted).
Owner earnings — the detail
Business quality
- ✓ ROIC exceeds the cost of capital (~10%)
- ✕ High incremental ROIC (compounding)
- ! CFROIC backs up the ROIC (79%, cash vs. accruals)
- ✓ Healthy balance sheet (low corporate debt)
- ✓ Durable competitive moat (multiple advantages)
Quality — cash · ROIC · reinvestment
- Strong, predictable cash generation. Reported operating cash flow is inflated by the float; the real adjusted free cash flow was US$1.13bn over the last twelve months and falls against the US$1.48bn of FY2025, depressed by investment. The cash is real, but the headline metric misleads.
- Sustainably high ROIC. The level is high (17.4%, above the 10% bar), but the incremental figure is negative (−1.1%): invested capital grew US$3.8bn in the year and operating income fell. That is what one expects in an investment phase and, at the same time, what remains to be demonstrated.
- Creates economic value (EVA). CFROIC (~12%) clears the 10% bar → positive economic profit (~+2pp on capital). It reinvests above the hurdle: it doesn't just earn, it creates value — the engine of compounding.
- Reinvestment runway. Wide: GMV can multiply given low regional penetration, TPV grows 56%, the book 75% and advertising is barely monetized. The runway exists; the return on reinvesting into it is what has to be verified year by year.
- Grows above its intrinsic g. Grows ~42% versus an intrinsic growth rate (reinvestment × ROIC) of ~7%: the gap is funded by float (negative working capital, funds payable ~US$13bn) and capital-light asset structure, without diluting. It's a structural funding advantage, with the caveat that the float is third-party funding (watch for a contraction).
- Accounting distortions to watch. (1) the float inflates operating cash flow; (2) the expected credit loss allowance front-loads losses and depresses accounting earnings; (3) the company reports by geography rather than by line → the segment margin is an analyst estimate, the largest uncertainty in the case.
Revenue trajectory
Values in US$ bn. The % over each bar is the year-over-year (YoY) growth — each year, historical and projected, vs the prior one (the TTM vs the TTM from a year ago). The path comes from the same source as the table; years without their own series in the model are interpolated between the anchors. Historical solid, projection in a lighter shade.
Where the growth comes from · by geography
Weight in revenue and year-over-year (YoY) growth, in reported USD.
Second quarter 2026 rates in dollars. The engine is Brazil and Mexico (~77% of revenue). At constant currency the ranking flips —Brazil +42%, Mexico +38%, Argentina +48%—: the difference between the two readings is foreign exchange, which this quarter blew in favor in Brazil and Mexico and against in Argentina.
Growth engine — operating drivers
Annual levels from the official filing (10-K); the % over each bar is the year-over-year (YoY) growth vs the prior year.
Each line breaks down into volume × price: commerce is GMV × take rate, payments are TPV × monetization and credit is book × interest margin. Annual fiscal-year levels; in the second quarter of 2026 GMV grew 44%, TPV 56% and the book 75%, while the take rate and the interest margin are moving lower by choice.
Projections
| Metric | FY23 | FY24 | FY25 | TTM | +1A | +2A | +3a | +4A | +5a |
|---|---|---|---|---|---|---|---|---|---|
Revenue | $15.1 bn | $20.8 bn (+38%) | $28.9 bn (+39%) | $35.2 bn (+46%) | $49.3 bn (+40%) | $66.2 bn (+34%) | $85.8 bn (+30%) | $106.8 bn (+24%) | $128.2 bn (+20%) |
Gross profit | $7.6 bn | $9.6 bn (+26%) | $12.9 bn (+34%) | $15 bn (+36%) | $20.2 bn (+34%) | $27.1 bn (+34%) | $36.5 bn (+34%) | $44.6 bn (+22%) | $54.5 bn (+22%) |
Gross margin | 50.2% | 46.1% (-4pp) | 44.5% (-2pp) | 42.7% (-3pp) | 42.6% (-0%) | 42.6% (-0%) | 42.5% (-0%) | 42.5% (+0%) | 42.5% (+0%) |
Income from operations | $2.2 bn | $2.6 bn (+19%) | $3.2 bn (+22%) | $2.9 bn (-2%) | $3.7 bn (+27%) | $5.7 bn (+54%) | $8.2 bn (+45%) | $11.3 bn (+37%) | $14.7 bn (+30%) |
Operating margin | 14.6% | 12.7% (-2pp) | 11.1% (-2pp) | 8.3% (-4pp) | 8.7% (+5%) | 9.1% (+5%) | 9.6% (+5%) | 10.5% (+9%) | 11.5% (+9%) |
Net income | $1 bn | $1.9 bn (+94%) | $2 bn (+5%) | $1.9 bn (-9%) | $2.7 bn (+43%) | $3.8 bn (+43%) | $5.4 bn (+43%) | $7.3 bn (+35%) | $9.9 bn (+35%) |
GMV | — | $51.5 bn | $65 bn (+26%) | $77.3 bn | $100.7 bn (+30%) | $131.3 bn (+30%) | $171 bn (+30%) | $207.6 bn (+21%) | $252 bn (+21%) |
Credit portfolio | — | $6.6 bn | $12.5 bn (+89%) | $16 bn | $22.9 bn (+43%) | $32.8 bn (+43%) | $47 bn (+43%) | $57.4 bn (+22%) | $70 bn (+22%) |
Owner earnings (adjusted FCF, ex-float) | — | $1.2 bn | $1.5 bn (+22%) | $1.1 bn | $1.6 bn (+46%) | $2.4 bn (+46%) | $3.5 bn (+46%) | $5.3 bn (+51%) | $8 bn (+51%) |
The % are the annual (year-over-year) growth: each year —historical and projected— vs the prior one; the TTM (trailing 12m) vs the TTM of a year ago, to avoid overlapping windows. The historicals are exact figures from the official filings; the projected years come from the year-by-year model (the intermediate years without their own series are interpolated between the anchors). The projected columns (+1y…+5y) are 12-month windows counted from the TTM close (30-jun-2026): the projection starts from the most recently reported data, not the fiscal year. The projected base is realistic and unbiased — the risk discount is applied at the end, via the required return. The rationale for each metric is in the (i).
Growth quality
- Real, not nominal, growth. The 46% in dollars is genuine: at constant currency the consolidated figure grows 43%, with Brazil +42% and Mexico +38%. The difference between reported and constant currency is foreign exchange, not Argentine inflation.
- Where the engine comes from. Concentrated and healthy: Brazil + Mexico ~77% of revenue; growth doesn't depend on Argentina (~18%, which additionally subtracts in USD).
- Realistic deceleration. The base starts at ~37% (the floor of the actual trajectory: FY24 +37.5%, FY25 +39.1%, TTM +42.1%) and decays to ~20% by year five — decelerating from the actual historical rate, without extrapolating the headline or starting below it.
- Driven by volume (not price/FX). GMV +44%, items sold +45%, TPV +56%, book +75%: growth comes from real volume (units, users, transactions). In fact price works against it, because the company is deliberately lowering it.
- Reinvestment runway. E-commerce and fintech penetration still low in LatAm + advertising still nascent → years of runway to reinvest at high ROIC. It's the engine of compounding.
- Capital cycle / competition. The sector attracts capital (Amazon, Shopee, Nubank) that pressures it; the moat (proprietary logistics, network) protects share, but defending it has a cost — part of the current margin compression.
Moat strength
The business and its moat
The business: what it does and how it makes money
MercadoLibre is Latin America's leading commerce and fintech ecosystem, operating in 18 countries with its value concentrated in Brazil and Mexico. The core is the marketplace —with a network effect between buyers and sellers— and around it revolve four mutually reinforcing lines: payments (Mercado Pago, on and off platform), credit (Mercado Crédito), in-house logistics (Meli Logistics) and advertising. Every transaction generates data that improves credit underwriting and ad targeting, which in turn drives more commerce: a self-reinforcing flywheel.
Because each line is different in nature, each is valued separately (sum of the parts): commerce and advertising are a capital-light, high-margin business; payments, a high-volume acquiring operation; and credit, a lending business funded by its own book. In FY2025 GMV reached ~$65bn (~$77bn over the last twelve months) and Mercado Pago's TPV ~$278bn (~$343bn), with revenue of ~$35bn over the last twelve months, growing 46% in dollars.
Scale and competitive position
MercadoLibre is the regional leader and has been gaining share. Growth is driven by Brazil (~53% of revenue, +59% in dollars in the second quarter of 2026) and Mexico (~24%, +55%); Argentina (~18%) decelerated to +20% in dollars. On a constant-currency basis the ranking flips —Brazil +42%, Mexico +38%, Argentina +48%—, which shows how much of the headline is foreign exchange: the consolidated figure grows 50% as reported and 43% at constant currency. Penetration of e-commerce and digital financial services in the region remains low versus developed markets, so the growth runway is secular rather than a cycle peak.
One financial trait sets the business apart: working capital is negative. The float of funds payable (~$13bn) funds operations at no cost, which lifts the return on invested capital and allows the company to grow faster than earnings alone would fund.
The moat: why it's hard to compete
The moat rests on four mutually reinforcing sources. First, the marketplace's network effect: more buyers attract more sellers, and vice versa, an equilibrium that's hard to dislodge. Second, proprietary logistics (Meli Logistics): years and billions of dollars invested in distribution and delivery centers, expensive and slow to replicate, that sustain delivery times an entrant can't immediately match. Third, the ecosystem's proprietary credit data: the transaction history of buyers and sellers lets the company lend with information a traditional bank doesn't have. Fourth, more than 25 years of brand and trust in a region where trust is scarce.
The combination of network, logistics, data, and brand is the barrier: a competitor — Amazon, Shopee, or Nubank — can attack one piece, but replicating all four at once, integrated, is truly difficult.
Moat direction and threats
The moat is wide and holds: the logistics network is gaining density, advertising and credit deepen the monetization of the same traffic, and Mercado Pago is advancing toward becoming the user's primary financial relationship (88 million fintech users, ~$23bn in assets under management, +68% year over year). The current investment phase —shipping subsidies, first-party inventory, credit and the card— is largely spending aimed at reinforcing that moat. The engagement metrics follow along —items per buyer are up 14% (19% in Brazil), active sellers 29%, and the share of users buying across three or more categories hit an all-time high—, but these are measures of volume and adoption: what is missing is the unit-economics gap, revenue per user against the cost to serve, that would prove the moat is widening.
The threats are concrete: competition from Amazon, Shopee and Nubank pressures prices and standards; the region's credit cycle could turn against the company if the book —growing 75% a year— expanded by loosening standards; and the float, today a no-cost funding advantage, depends on confidence and on fintech regulation —if it contracts, negative working capital reverses—.
Business / sector quality
- Relevant in 10 years / recurrence. Yes — e-commerce and payments are recurring habits; LatAm penetration ~half of US/UK → a long runway.
- Market share / cost. Regional e-commerce and fintech leader, gaining share; logistics scale lowers cost per shipment versus competitors.
- Pricing power. Real in advertising and financial services, but today it is used in reverse: the company subsidizes price —free shipping, discounts for paying with PIX, reduced seller take rates— to win volume and engagement.
- Operating leverage. High once mature (asset-light marketplace), but latent today: margin is compressed by voluntary investment, not yet realized.
- Supply-side / capital cycle. The sector attracts capital (Amazon, Shopee, Nubank) → pressuring margins; MELI defends with structural moats, but the cost of defending share is real.
- Behavior in a recession. E-commerce is defensive (everyday consumption), but credit is procyclical (delinquency rises in a LatAm recession). Mixed.
Solvency margin
Each pillar between danger and solid — the further right, the more room.
Reading for a float business: a current ratio near 1 is structural (not a warning sign), and net cash is float-adjusted (excludes customer cash).
Net cash position
Float-adjusted corporate position: gross cash is intertwined with customer float (third-party money), which is not the business's own surplus — this net figure is what counts.
Debt composition
Not all debt is equal: only the structural needs refinancing; the rest is operational (self-liquidating).
Only ~16% is structural debt (bonds, well spaced out); the rest is operating funding matched with the loan book — it does not need refinancing. Net cash covers the structural portion.
Company health / solvency
- ✓Leverage (net debt / EBITDA)Net cash $2 bn
- ✓Interest coverage (EBIT / interest)13.7x
- ✓Liquidity (current ratio)1.12×
- !Cash quality (CFROIC vs ROIC)CFROIC backs 79% of ROIC
- ✓Value creation (ROIC − 10% bar)+7pp
- ✕Malinvestment test (capex vs incremental ROIC)Investment mode 1.5x with incremental ROIC < cost of capital
- !Float / working capitalConsumes cash $0.7 bn (positive WC)
- ✓Dilution (SBC % of revenue + shares)SBC 0.8% of revenue
A traffic-light interpreted by the method (not generic): float (negative WC) adds up, capex is judged by incremental ROIC (malinvestment test), and a lender is not subjected to corporate solvency. The (i) shows the derivation of each number.
ⓘ Has a loan book: credit health (delinquency, coverage, normalized CECL, RoA/RoE) is evaluated separately, not with corporate solvency.
Health — balance sheet risks
- Borrowed float (contraction risk). Funds payable (~US$13bn) is third-party funding, not owned capital; a contraction (fintech regulation, higher rates, a run on trust) reverses negative working capital and that funding advantage evaporates.
- FX / country exposure on the balance sheet. ~95% of revenue and a good share of assets in currencies that devalue (real, peso); capital controls in Argentina. It's a macro risk that solvency ratios don't capture.
- The balance sheet carries the credit book. The book (+75%) adds procyclical risk: in a regional recession delinquency rises and the expected credit loss allowance may understate next-cycle losses. It is assessed separately (return on assets and on equity, delinquency, coverage).
- No maturity wall. Low corporate debt (~US$1.75bn in 2026/2031/2033 bonds), well spaced out; net cash covers it. No pressing refinancing risk (the rest of the funding is operational, matched against the credit book).
- Structural liquidity (float). A current ratio near 1 is structural (funded by float/payables), not a red flag; and net cash is float-adjusted (excludes customer money).
Who runs it
- Orderly transition: the founder moved to Executive Chairman and promoted a long-tenured internal CEO — continuity, not rupture.
- Long-term orientation: 28 consecutive quarters of revenue growth above 30%.
- Communicates the margin/growth trade-off candidly (quantifies the 5-6pp margin drag).
Capital allocation — indicators
The quality test: it reinvests at -1% vs the 10% bar (market opportunity cost) → each $1 retained creates value.
Sources and uses of cash
How cash comes in and how it is deployed. In green, the business's own cash (the owner-FCF it generates and reinvests); in gray, the float and credit — customer and funding money, which is not the shareholder's.
The cash the business generates on its own (adjusted free cash flow of ~$1.13bn over the last twelve months) is reinvested in full: 100% retained, with no dividend and no buyback. The reinvestment runway exists, but the return on new capital is not yet visible —the incremental figure on invested capital is negative—, so the decision to retain everything is judged by the maturing of the cohorts rather than by today's result. Separately, the float and the funding are redeployed into the credit book (third-party money, not shareholder earnings): that is the float machine, not allocation of the company's own capital.
Shares — ownership and dilution
Who owns the shares — the alignment and whether there is a controlling shareholder.
Minimal dilution: SBC represents less than 2% of value per year and the share count is ~flat — it does not erode value per share.
Management / capital allocation
- Capital allocation (ROIC). Reinvests aggressively at high ROIC (incremental ~30%), funded by float (negative working capital, funds payable US$13bn). Pays no dividend — correct, given a reinvestment runway.
- Alignment / skin in the game. Orderly transition: Galperín (founder, 1999) moved to Executive Chairman in Jan-2026 and promoted an internal CEO (Ariel Szarfsztejn); the Galperin Trust owns ~7% (2nd-largest shareholder after Baillie Gifford), and compensation is tied to KPIs. Long-term orientation.
- Integrity / communication. Communicates the margin/growth trade-off candidly (quantifies the 5-6pp drag). SBC contained (~1% of revenue) → doesn't inflate FCF the way it does at other tech companies.
- Acquisitions / diworsification. Growth is mostly organic rather than through serial acquisitions; the bets (first-party inventory, cross-border trade, the card, groceries) are extensions of the core business.
Why it is cheap
- Margin compression: operating income fell 17% year over year in the second quarter of 2026 and the margin came in at 6.7%. The market extrapolates the decline and sells, reading a voluntary investment phase as structural deterioration. The stock trades under pressure.
- Misleading headline multiple: 50× P/E on margins and cash depressed by investment. Automated screens flag it as expensive precisely while it is investing, which is the wrong lens.
- Expected credit losses: the lifetime expected loss allowance (US$4.25bn on the June 2026 balance sheet, versus US$3.06bn in December) front-loads losses and depresses accounting earnings while the book grows 75%. The company spells it out: netting credit revenue against provisions, the share of consolidated revenue was slightly higher than a year earlier.
- Irregular accounting earnings from foreign exchange (−US$39M in the first half of 2026, versus −US$172M a year earlier) and taxes.
- The lending business (Mercado Crédito) is hidden inside an e-commerce company and is hard to value, so many investors assign it no value (it's a free optionality).
The discount comes mostly from optics —margin compression and a misleading multiple on depressed numbers— rather than from deteriorating demand: growth accelerated to 46% in dollars. The caveat, which is worth not minimizing, is that the compression has already reached last-twelve-month operating income and the incremental return on capital: the market is not discounting something that did not happen, but betting that it does not reverse.
Return asymmetry — risk/reward
The annual return (CAGR at 5 years) in each scenario, with the total period return below — the margin of safety made visual: upside range wide, downside range narrow.
Even in the bear scenario, the return holds at -6%/year (-25% total): the margin of safety protects the downside. The bull (+28%/year, +250% total) exceeds it comfortably — a favorable asymmetry, with a narrow downside range and a wide upside range.
Bear case — disconfirmation
- The operating margin does not recover: the investment is not optional but the structural cost of defending against Amazon, Shopee and Nubank, so the 7% to 8% range is the ceiling and not a temporary floor. Evidence for this reading already exists: the incremental return on invested capital is negative, with US$3.8bn of new capital in the year and lower operating income. With the margin stabilizing at ~8.5% the annual return is -6%.
- Credit is growth bought by loosening standards: the book grows 75% while the mix shifts toward the card (from 43% to 47%), whose net interest margin after losses is already negative (−2.5% versus breakeven a year earlier). The company attributes this to new cohorts being dilutive at first and maturing later; if they do not mature, the credit optionality turns into a hole.
- The float is borrowed, not owned: if funds payable to customers (US$13.1bn of restricted cash as of June 2026) contract because of fintech regulation, rising rates or a loss of confidence, negative working capital reverses and that funding advantage evaporates.
Bull case — the thesis for
- The margin recovery is a matter of when, not whether: the compression to 6.7% is voluntary investment rather than structural erosion, and the company already demonstrated the pattern when it cut seller shipping charges in 2024 and 2025. The early evidence is there: free and slow shipments already deliver positive variable contribution across half of the price ranges between 19 and 79 reais, active sellers are growing 29% and items per buyer 14%. On a revenue base 46% larger, every point of recovered margin is worth a great deal.
- Credit is a second earnings engine the market discounts as risk: Mercado Crédito is scaling its loan book at a high RoA that, even normalizing CECL to steady-state charge-offs, sustains attractive returns. Valued as a lender (P/E on normalized earnings), it adds value rather than subtracting it.
- Advertising is the hidden margin lever: it grows at a very high margin and is barely monetized relative to GMV, so it lifts the consolidated margin without additional investment in fixed assets.
- The secular runway remains intact: penetration of e-commerce and digital financial services in Brazil and Mexico is still low versus developed markets. Engagement metrics are at record levels —assets under management per user reached US$264 (+29%) and the ratio of daily to monthly active users hit an all-time high—, and the in-house logistics network reinforces the moat.
Risks — what breaks the base case
- Operating margin doesn't recover. Operating income fell 17% year over year in the second quarter of 2026 (margin 6.7%) and the last-twelve-month figure is already lower than a year ago. The base case depends on a recovery to ~11.5%; if it does not happen, the adverse scenario prevails. It is the risk of the case.
- Credit cycle (CECL / delinquency). Book +75% and an expected credit loss allowance of US$4.25bn on the balance sheet, versus US$3.06bn six months earlier. Net interest margin after losses fell to 20.7% and the card's to −2.5%. Watch delinquency in a regional recession.
- FX / Argentina. FX losses of −US$337M in FY25; inflation/devaluation distort the nominal figures (look at FX-neutral); capital controls.
- Competition and execution. Amazon, Shopee and Nubank apply pressure; and there is execution risk on the bets (first-party inventory, cross-border trade, groceries, digital banking).
- Demanding multiple. 50× P/E on depressed numbers: the entire case rests on the margin recovery; if it does not arrive, the multiple recomposes against cash.
Lenses — the value investing thinkers
Each thinker's analytical framework applied to our data.
A quality business at a reasonable price; the margin is limited.
- Buffett / Graham Quality + margin of safety
A wide moat and ROIC 17% above the 10% bar, with a +40% margin → a quality business at a good price.
- Peter Lynch Growth at a reasonable price (GARP)
A fast grower growing 18% at a multiple/growth of 2.6 → expensive for its growth.
- Joel Greenblatt Cheap and high-return (Magic Formula)
Earnings yield 3% (EBIT/EV) + ROIC 17% → falls outside the Magic Formula.
- Howard Marks Perception vs reality + cycle
The price discounts 13% vs our 18%: perception is more pessimistic than reality, with an identified source of the discount.
- Seth Klarman Capital protection (bear scenario)
Bear-scenario floor -6%/yr, bull-scenario ceiling +28%/yr over 5y: reasonable asymmetry, without an ample cushion.
- Pat Dorsey Moat strength (Five Rules)
A wide moat, stable; sources: network effects, cost advantage, intangibles, switching costs, efficient scale → passes the Five Rules.
- Aswath Damodaran Expectations implied by the price
Justifying the price requires discounting 13%, within what we project (18%) — the story squares with the numbers.
