A Full Year Where the Underlying Business, Not a Currency Reversal, Actually Drove the Improvement
For the first time in several years of this coverage, MercadoLibre closed a fiscal year without a single-quarter Venezuelan currency shock dominating the full-year read. FY2016 net revenue grew 29.6% to $844.4 million, the fastest full-year growth rate since FY2013, and net income grew 28.9% to $136.4 million - both accelerating from FY2015's 17.1%/45.5% pair, rather than snapping back from a prior year's one-off charge the way FY2015 rebounded from FY2014's Venezuela impairment. There was still a Venezuela-related pretax charge inside the year (a Q2 2016 hit the tax-rate footnote references only in passing, smaller than 2015's), but it didn't dominate the annual numbers the way SICAD/SIMADI shocks did in 2014 and 2015.
The growth is real, but it isn't even. Brazil (53.9% of revenue) grew 56.6% and gained 2.6 points of direct-contribution margin; Argentina (31.1% of revenue) grew a more modest 7.0% and gave back 3.0 points of margin. Mexico and Venezuela - together under 10% of revenue - each lost more than six points of margin. Consolidated direct-contribution margin barely moved (37.6% vs. 38.2% a year ago, or 39.2% vs. 40.8% stripping out the recurring Venezuela impairment) - a full year where the headline looks stable specifically because Brazil's improvement offset everyone else's deterioration, not because the whole portfolio is executing well simultaneously. See Five Country Segments below.
The Prescription
MercadoLibre should keep concentrating capital and management attention on Brazil and Argentina, which together generated 85% of FY2016 revenue and did essentially all of the year's margin improvement - Brazil alone added 2.6 points of direct-contribution margin on 56.6% revenue growth, a combination no other segment came close to. The MercadoPago payments business is the clearest lever inside that: total payment volume grew 49.6% (faster than net revenue), and TPV as a share of GMV kept climbing, meaning a growing share of the platform's own economics now runs through a business MercadoLibre fully controls rather than through third-party payment rails.
What it should stop doing: continuing to run Mexico and Venezuela as if they were still meaningful profit contributors when the numbers say otherwise. Mexico's direct-contribution margin fell 10.9 points to just 11.6% and Venezuela's fell 6.7 points to 15.4% (or roughly 10 points on an ex-impairment basis) - combined, the two segments now generate barely $11 million of direct contribution on $83.5 million of revenue, about 3.5% of the consolidated total, while still requiring separate country-level infrastructure, local banking relationships, and - in Venezuela's case - a third round of currency-regime disclosure. Two segments this small and this marginal deserve either a genuine turnaround plan disclosed to investors or a candid acknowledgment that they're being carried for optionality, not profit.
Key Financial Metrics
Year ended December 31, 2016 vs. year ended December 31, 2015 - consolidated, reported in USD (MercadoLibre reports natively in dollars; no FX conversion is needed). MercadoLibre does not disclose an Adjusted EBITDA figure in this filing.
| Metric | FY2016 | FY2015 | YoY |
|---|---|---|---|
| Net revenues | $844.4M | $651.8M | ✅ +29.6% |
| Cost of net revenues | $(307.5)M | $(215.0)M | ⚠️ +43.0% - taxes on revenue rose to 9.0% of net revenues (from 8.1%) and MercadoPago collection fees grew with payment-volume penetration |
| Gross profit (63.6% margin) | $536.9M | $436.8M (67.0% margin) | ✅ +22.9%, margin -3.4pp |
| Product and technology development | $(98.5)M | $(76.4)M | ⚠️ +28.9% - roughly tracking revenue growth for the first time in several years |
| Sales and marketing | $(156.3)M | $(128.6)M | ⚠️ +21.5% - slower than revenue growth |
| General and administrative | $(87.3)M | $(76.3)M | ⚠️ +14.4% - slower than revenue growth |
| Impairment of long-lived assets | $(13.7)M | $(16.2)M | ✅ -15.4% - still entirely Venezuela real estate, but a smaller charge than FY2015's |
| Income from operations (21.4% margin) | $181.1M | $139.2M (21.4% margin) | ✅ +30.1%, margin flat |
| Interest income and other financial gains | $35.4M | $20.6M | ✅ +71.8% - larger invested cash balance |
| Interest expense and other financial charges | $(25.6)M | $(20.4)M | ⚠️ +25.5% - convertible-note discount accretion continuing on schedule |
| Foreign currency (losses) / gains | $(5.6)M | $11.1M | 🔴 Swung from a gain to a loss |
| Blended tax rate | 26.4% | 29.7% | ✅ -3.3pp |
| Effective tax rate | 30.0% | 30.5% | ✅ -0.5pp |
| Net income | $136.4M | $105.8M | ✅ +28.9% |
| Diluted EPS | $3.09 | $2.40 | ✅ +28.8% |
| Free cash flow» | ~$121.4M | n/a (FY2015 not restated in this filing) | Operating cash flow $190.3M less ~$68.9M of property/equipment and intangible-asset purchases |
| Cash, ST & LT investments (period end) | $641.3M | $556.6M | ✅ +15.2% |
| Loans payable and other financial liabilities (period end) | $313.5M | $296.3M | ⚠️ +5.8% - almost entirely the $330M convertible notes, net of discount amortization |
Net income growth (28.9%) tracked revenue growth (29.6%) almost exactly this year - a genuinely different shape from FY2014's Venezuela-driven collapse or FY2015's post-impairment rebound. Operating margin held flat at 21.4% even as gross margin compressed 3.4 points, because operating-expense growth across all three expense lines ran slower than revenue for the first time in several years of this coverage.
Key Operational Metrics
Year ended December 31, 2016 vs. year ended December 31, 2015
| Metric | FY2016 | FY2015 | YoY |
|---|---|---|---|
| Gross merchandise volume» (GMV, excludes motor vehicles/vessels/aircraft/real estate) | $8,048.1M | $7,150.8M | ✅ +12.5% |
| Total payment volume» (TPV) | $7,753.7M | $5,184.1M | ✅ +49.6% |
| Total payment transactions | 138.7M | 80.4M | ✅ +72.5% |
| Confirmed registered users (cumulative, period end) | 174.2M | 144.6M | ✅ +20.4% |
| New confirmed registered users (in period) | 29.5M | 23.7M | ✅ +24.5% |
| Items sold | 181.2M | 128.4M | ✅ +41.0% |
| Items shipped | 86.5M | 45.2M | ✅ +91.4% |
| Unique buyers | 27.7M | 23.6M | ✅ +17.4% |
| Take rate» (net revenues / GMV) | 10.49% | 9.12% | ✅ +1.37pp |
Items sold growing 41.0% while GMV grew only 12.5% - a 28.5-point gap - continues the pattern this coverage has tracked since 2014: a growing share of transactions are lower-average-value items, and currency effects keep understating dollar-denominated GMV relative to actual platform activity. TPV growing 49.6%, nearly four times GMV's growth rate, is the more important number here - MercadoPago is now processing payment volume equivalent to 96.3% of GMV (up from 72.5% a year ago), meaning MercadoLibre's own payments rail has become the default way value moves across the platform rather than a bolt-on option.
Five Country Segments - Brazil Carried the Year; Mexico and Venezuela Both Lost More Than Six Points of Margin
MercadoLibre reports the same five geographic segments adopted in Q3 2010 - Brazil, Argentina, Mexico, Venezuela, and Other Countries.
| Segment | FY2016 Revenue | YoY Revenue | Direct contribution margin, FY2016 | Direct contribution margin, FY2015 | Margin change |
|---|---|---|---|---|---|
| Brazil | $455.0M | ✅ +56.6% | 40.5% | 37.9% | ✅ +2.6pp |
| Argentina | $262.3M | ✅ +7.0% | 42.0% | 45.0% | 🔴 -3.0pp |
| Mexico | $46.3M | ✅ +14.9% | 11.6% | 22.5% | 🔴 -10.9pp |
| Venezuela | $37.2M | 🔴 -8.1% | 15.4% | 22.1% | 🔴 -6.7pp (ex-impairment ≈52.2% vs. ≈62.2% a year ago, ⚠️ -10.0pp) |
| Other Countries | $43.6M | ✅ +23.3% | 27.6% | 30.4% | 🔴 -2.8pp |
| Total | $844.4M | ✅ +29.6% | 37.6% | 38.2% | ⚠️ -0.6pp (ex-impairment 39.2% vs. 40.8%, ⚠️ -1.6pp) |
Brazil, now 53.9% of consolidated revenue (up from 44.6% a year ago), is doing the heaviest lifting in this business by a wide margin - 56.6% revenue growth and a margin gain, a combination no other segment matched. Argentina's revenue growth slowed sharply to 7.0% (from FY2015's much faster pace) as the peso's continued depreciation ate into dollar-reported growth, and its margin gave back three of the eight points it gained in FY2015. Mexico's ten-point margin decline is the steepest of any segment this year and isn't currency-explained the way Venezuela's is - the filing attributes it to higher pre-tax losses from a peso devaluation flowing through operating costs, not a one-time item. Venezuela's headline decline is smaller than it looks because the entire $13.7 million impairment sits inside its $5.7 million of direct contribution; stripping it out, the segment's operating margin still fell 10 points, a genuine deterioration on top of the accounting noise, on a revenue base that's now shrinking outright (-8.1%).
Beyond the Usual
The stock closed the year already above the price at which its own convertible-note dilution hedge stops protecting the Company
MercadoLibre's $330 million of 2.25% convertible senior notes due 2019 became convertible at holders' option in Q3 2016, after the stock traded above 130% of the notes' $126.02 conversion price for the required window. To limit dilution from a rising stock price, the Company had separately bought capped-call transactions in 2014 with a $126.02 strike and a $155.78 cap price - meaning the hedge fully offsets dilution only up to that cap, and provides zero additional protection above it. MercadoLibre's stock closed FY2016 at $156.14, already $0.36 above that cap, and the 10-K itself discloses the notes' if-converted value exceeded their principal by $78.8 million at year-end.
This isn't a hypothetical risk to flag for the future - by the filing's own year-end numbers, the capped-call hedge has already stopped providing incremental dilution protection at the exact price the stock closed the year at. Any further appreciation in MercadoLibre's own stock price between now and the notes' July 2019 maturity increases shareholder dilution on a one-for-one basis, with no offsetting hedge cushioning it, the opposite of what a rising stock price usually means for existing shareholders holding convertible-note-financed companies.
A newer, higher-value tranche of executive retention compensation is now indexed directly to MercadoLibre's own stock price
The 2016 Long-Term Retention Plan, adopted August 2, 2016, layers a sixth overlapping stock-price-linked award on top of the 2009-2015 tranches this coverage has tracked since 2013 - and it's the largest yet: executive officers are set to receive approximately $13.4 million under the 2016 plan alone (versus $10.3 million for 2015's tranche, $8.3 million for 2014's), with the variable portion's payout scaling directly with the ratio of MercadoLibre's own average year-end stock price to a fixed $111.02 baseline (the average close during the final 60 trading days of 2015).
Each year's tranche size has grown roughly in step with the stock's own multi-year rally, which means compensation expense under this plan mechanically inflates further whenever the stock keeps climbing - the same dynamic this coverage has flagged pushing G&A growth in prior periods, now compounding across six simultaneously-vesting plans rather than one or two.
MercadoLibre disclosed two small software-company acquisitions in 2016, both aimed at internal capability rather than revenue
In February 2016, the Company acquired Monits S.A., an Argentine software developer, for $3.1 million; in June 2016, it acquired Axado, a Brazilian logistics-software company serving e-commerce shippers, for $5.5 million. Both are explicitly framed as capability acquisitions (software development generally, and transportation-management-system/shipping technology specifically) rather than revenue-generating bolt-ons, a different rationale from the real-estate-classifieds acquisitions (Chile/Mexico's Guia de Inmuebles, Mexico's Metros Cúbicos) this coverage has tracked in recent years.
The Company took on four new local-currency credit lines in 2016, each in a different country, none disclosed as a coordinated financing strategy
Separate from the $330 million convertible notes, MercadoLibre's Brazilian subsidiary drew four lines of credit from Brazil's national development bank (BNDES) to fund equipment purchases; its Uruguayan subsidiary drew a working-capital line from Citibank; and its Chilean subsidiary drew a working-capital line from Banco de Chile - on top of the pre-existing Argentine equipment line and Venezuelan mortgage already disclosed in prior filings. Individually each is small (the largest, Chile's, was $9.6 million outstanding at year-end), but the pattern - a different local bank relationship opened in a different country nearly every year since 2013 - suggests working-capital and equipment financing across Latin America's more currency-controlled economies is becoming a recurring operational necessity, not a one-off financing choice.
The São Paulo tax claim remains active judicial litigation with zero reserve, now in its ninth consecutive year unresolved
The 2007-2010 São Paulo municipal tax claim - contesting taxes and fines the Company still calls "remote" risk despite years of adverse administrative rulings - remains before the 9th Treasury Court of the City of São Paulo with no ruling yet, the same posture disclosed in every filing since November 2013. A separate, smaller 2005-2007 claim sits in an identical unresolved posture before the 8th Public Treasury Court. Two new, smaller Brazilian tax administrative claims also surfaced in late 2016 (a $945 thousand PIS/COFINS credit dispute in November, a $3.7 million version in December), both still at the administrative-defense stage.
This is a genuinely interesting footnote pattern rather than a governance concern on its own: MercadoLibre has now carried multi-year, multi-million-dollar Brazilian tax exposure completely unreserved for the better part of a decade, on the same "remote risk" legal-counsel opinion each time, even as the claims have moved through years of adverse and mixed rulings without ever actually being resolved.
Coverage Table
| Theme | FY2016 | FY2015 | YoY | Why it matters |
|---|---|---|---|---|
| Net revenue | $844.4M | $651.8M | ✅ +29.6% | Fastest full-year growth since FY2013, and accelerating rather than rebounding from a prior-year shock |
| Net income | $136.4M | $105.8M | ✅ +28.9% | Tracked revenue growth almost exactly - a genuinely different shape from FY2014/FY2015's Venezuela-driven swings |
| Total payment volume (TPV) | $7,753.7M | $5,184.1M | ✅ +49.6% | Now equivalent to 96.3% of GMV - MercadoPago is the default payment rail on the platform |
| Consolidated direct-contribution margin (ex-impairment) | 39.2% | 40.8% | 🔴 -1.6pp | Brazil's gain offset Mexico's and Venezuela's losses - a stable headline masking uneven segment performance |
| Convertible-note capped-call cap price vs. year-end stock price | Stock $0.36 above the $155.78 cap | n/a | 🔴 | The dilution hedge stopped providing incremental protection right at the price the year closed |
Target Valuation Range
DCF fair-value range: roughly $4,025 million (base case) to $8,071 million (bull case) enterprise value, against a $6,567.0 million actual enterprise value - back inside the fair-value band after Q3's overshoot. Base case: overvalued relative to trailing cash flow, though the gap has narrowed from most of this coverage's history; bull case now clears enterprise value outright for the first time since FY2014, reflecting a full year where growth genuinely accelerated rather than merely rebounding.
The stock closed FY2016 at $156.14 on December 30, 2016 (MercadoLibre has never split its common stock since its August 2007 IPO, so no split adjustment applies).
| Market cap → enterprise value | FY2016 |
|---|---|
| Share price (period-end) | $156.14 |
| Shares outstanding | 44,157,364 |
| Market capitalization | $6,894.7 million |
| Plus: loans payable | $313.5 million |
| Less: cash & investments | $641.3 million |
| Enterprise value | $6,567.0 million |
Enterprise value is up 27.6% from FY2015's ~$5,148 million (derived from that filing's own balance sheet), split between the stock's rally and the growing cash and debt balances.
| Valuation multiples | FY2015 | FY2016 | Change |
|---|---|---|---|
| Enterprise value | ~$5,148 million | $6,567.0 million | ⬆ up |
| EV/Sales | n/a | 7.78x | ⬇ down (revenue grew faster than EV) |
| P/E (TTM) | n/a | 50.5x | - |
| EV/FCF (TTM) | n/a | 54.1x | - |
Both multiples are still expensive by conventional standards, though less extreme than several of this coverage's Venezuela-charge-distorted prior readings.
| Scenario | Key assumption | Implied EV | % of actual EV |
|---|---|---|---|
| Current (FY2016 close) | actual market price | $6,567.0 million | 100% |
| Base | Revenue growth 20%→10% over 5yrs; FCF margin 14.4%→28%; 13% discount, 4% terminal growth | $4,025 million | 61.3% |
| Bull | Revenue growth 24/20/17/14/11%; FCF margin 22.8%→34%; 11% discount, 5% terminal growth | $8,071 million | 122.9% |
This is the first time in several years of this coverage the bull case has cleared actual enterprise value by a meaningful margin, rather than falling just short or just barely clearing it.
Reverse DCF: holding the 13% discount rate and solving for the perpetual FCF growth rate that would justify today's $6,567.0 million enterprise value on the $121.4 million trailing FCF base gives roughly 11.0% growth, forever - back up from the low-10%/high-9% readings this coverage measured in early-to-mid 2015, as the year's stock rally outpaced FCF growth.
Whether Mexico and Venezuela's margin erosion stabilizes or keeps widening, and whether the already-breached capped-call ceiling starts mechanically diluting shareholders if the stock keeps climbing, are the two clearest swing factors for this valuation gap going into 2017.
MercadoLibre, Inc.'s Annual Report on Form 10-K for the fiscal year ended December 31, 2016, filed with the U.S. Securities and Exchange Commission and signed February 24, 2017. Historical MELI share price data covers month-end closes from December 2014 through December 2016; MercadoLibre has never split its common stock since its August 2007 IPO, so no split adjustment applies to these figures. No presentation, press release, or transcript was located for this filing, so this post does not include a management-commentary section.