MELI The First Full-Year Loss Since 2007 - What the Footnotes Say About Why

The Full-Year Verdict on a Deliberate Margin Trade - and a Sequential Turn in Q4

MercadoLibre's FY2018 10-K puts a final number on the growth-for-margin trade this coverage has tracked across Q2 and Q3: full-year net revenue grew 18.3% to $1,439.7 million, but the Company posted a $36.6 million net loss - its first full-year net loss this coverage has recorded, reversing $13.8 million of net income in 2017 (restated for ASC 606). Free-shipping subsidies, the single biggest driver of the story all year, nearly tripled to $424.8 million from $181.6 million, and gross margin fell to 48.4% from 59.2%. Sales-and-marketing expense grew even faster (up roughly 48% for the year) as buyer protection program costs, marketing spend, and bad debt all scaled with the business.

But the annual number masks a real sequential turn that the quarterly filings, read individually, only hinted at. Q4 2018 standalone - derived by subtracting the nine-month figures already reported in Q3's 10-Q from the full-year total - posted revenue of $428.0 million (+19.5% YoY), an operating loss of just $0.8 million, and a net loss of just $2.3 million, a dramatic sequential improvement from Q2's $57.7 million operating loss and Q3's $10.1 million standalone net loss. The worst of the margin compression happened in the first half of the year, not the fourth quarter - whether that's the start of a genuine stabilization or simply a seasonally stronger quarter (Q4 typically benefits from holiday-season e-commerce demand across MercadoLibre's markets) is the single most important open question this filing leaves for the next one to answer.

Underneath both halves of the story sits a currency problem that got worse, not better, over the year: GMV growth decelerated sharply to just 6.4% for the full year (from Q3's already-currency-affected quarterly declines), even as total payment volume grew 34.4% and confirmed registered users grew 26.2% - the widest gross-versus-net divergence this coverage has measured from MercadoLibre, driven by a 50.5% Argentine peso devaluation and a 14.6% Brazilian real devaluation over the year. A U.S.-dollar-denominated GMV figure growing at barely a quarter of the pace of registered-user growth is a currency story more than a demand one, but it's also exactly the kind of gap a reader relying only on the headline GMV number would miss.

The Prescription

MercadoLibre should treat Q4's near-breakeven result as a hypothesis to prove out, not a victory to declare - the Company should explicitly disclose, in the next filing, whether Q4's improvement reflects a genuine stabilization in the shipping-subsidy and marketing spend that drove the first three quarters' losses, or simply the seasonal lift every e-commerce platform in these markets gets in the fourth quarter. Given that this year's entire net loss was effectively booked in the first nine months, a reader deserves to know which explanation is closer to true before assuming 2019 will look like a clean continuation of Q4's trend.

What it should stop doing: continuing to retire in-the-money convertible debt at a real premium to principal without quantifying the cost plainly in the MD&A. Retiring $263.7 million of principal amount of the 2019 Notes in August cost $348.1 million in cash plus 1,044,298 shares of common stock (see Beyond the Usual) - a genuine economic cost of having issued convertible debt with a conversion price the stock later blew through, and one that's disclosed in the notes but not framed anywhere as "this is what being in-the-money on old debt costs us." A company this financially sophisticated should be able to explain that tradeoff to shareholders directly rather than leaving it to be reconstructed from footnote arithmetic.

Key Financial Metrics

Year ended December 31, 2018 vs. year ended December 31, 2017 - consolidated, reported in USD (MercadoLibre reports natively in dollars; no FX conversion is needed); FY2017 figures reflect the Company's own ASC 606 retrospective restatement

Metric	FY2018	FY2017	YoY
Net revenues	$1,439.7M	$1,216.5M	✅ +18.3%
Cost of net revenues	$(742.6)M	$(496.9)M	🔴 +49.4% - shipping subsidies (+$243.2M) and collection fees
Gross profit (48.4% margin)	$697.0M	$719.6M (59.2% margin)	🔴 -3.1%, margin -10.8pp
Sales and marketing	$(482.4)M	$(325.4)M	🔴 +48.3%
General and administrative	$(137.8)M	$(122.2)M	⚠️ +12.7%
Product and technology development	$(146.3)M	$(127.2)M	⚠️ +15.0%
(Loss) income from operations	$(69.5)M	$56.3M	🔴 Swung from profit to loss
Foreign currency gains (losses)	$18.2M	$(21.6)M	✅ Favorable swing, mostly Argentine highly-inflationary-accounting mechanics
Net (loss) income	$(36.6)M	$13.8M	🔴 First full-year net loss this coverage has recorded
Net cash provided by operating activities	$230.9M	$269.0M	🔴 -14.2%
Free cash flow» (operating cash flow less capex, intangibles, property advances)	$133.2M	$194.1M	🔴 -31.4%
Cash, cash equivalents & restricted cash (period end)	$464.7M	$388.3M	✅ +19.7%, despite the $880M debt raise - most of the new cash was parked in short/long-term investments instead (see below)
Short-term and long-term investments (period end)	$737.7M	$244.2M	✅ +202.1%, includes a $284.3M Central Bank of Brazil mandatory guarantee (see Beyond the Usual)
Loans payable and other financial liabilities (period end)	$735.2M	$368.4M	🔴 +99.6%, mostly the new 2028 Notes net of the retired 2019 Notes
Q4 2018 standalone (derived): revenue / operating loss / net loss	$428.0M / $(0.8)M / $(2.3)M	$358.1M / n/a / n/a	✅ +19.5% revenue; near-breakeven, a sharp sequential improvement from H1/Q3

Adjusted EBITDA is not disclosed by MercadoLibre in this filing as a distinct non-GAAP measure, so it is omitted here rather than estimated.

Key Operational Metrics

Year ended December 31, 2018 vs. year ended December 31, 2017

Metric	FY2018	FY2017	YoY
Gross merchandise volume» (GMV)	n/a (dollar figure not disclosed; growth rate only)	n/a	⚠️ +6.4% - sharply decelerated, mostly currency devaluation
Total payment volume» (TPV)	n/a (dollar figure not disclosed; growth rate only)	n/a	✅ +34.4%
Confirmed registered users (cumulative, period end)	n/a (dollar figure not disclosed; growth rate only)	n/a	✅ +26.2%
TPV as % of GMV	90.2%	81.9%	✅ +8.3pp
Items shipped via MercadoEnvios as % of items sold	66.3%	55.8%	✅ +10.5pp
Free-shipping subsidies (netted against revenue)	$424.8M	$181.6M	🔴 +133.9%
Argentine Peso devaluation (full year)	~50.5%	n/a	🔴 Direct driver of GMV/net-revenue divergence
Brazilian Real devaluation (full year)	~14.6%	n/a	⚠️

GMV growing just 6.4% for the full year, against TPV's 34.4% and confirmed-registered-users' 26.2%, is the widest version yet of the gross-versus-net gap flagged in Q2 and Q3 - but this time the primary driver is currency, not the shipping-subsidy mechanic that dominated H1. A dollar-denominated GMV figure this depressed relative to underlying activity is a genuine disclosure challenge for a reader: local-currency GMV growth (not separately quantified at the consolidated level in this filing) is almost certainly running well ahead of the 6.4% headline number once the peso and real devaluations are stripped out.

Four Country Segments - Brazil's Margin Nearly Halved for the Year, Mexico Stayed the Weakest Link

MercadoLibre reports the same four geographic segments as this year's quarterly filings - Brazil, Argentina, Mexico, and Other Countries (Venezuela deconsolidated effective December 1, 2017).

Segment	FY2018 Revenue	YoY Revenue	Direct contribution margin, FY2018	Direct contribution margin, FY2017	Margin change
Brazil	$866.2M	✅ +25.4%	11.95%	31.73%	🔴 -19.78pp
Argentina	$376.6M	✅ +4.8%	32.40%	39.94%	⚠️ -7.54pp
Mexico	$109.1M	✅ +112.5%	-50.91%	-109.23%	✅ +58.32pp - still deeply negative, but the sharpest full-year improvement of any segment this coverage has measured
Other Countries	$87.8M	✅ +44.6%	9.38%	12.38%*	🔴 -3.00pp
Total	$1,439.7M	✅ +18.3%	12.38%	21.94%**	🔴 -9.56pp

*FY2017's "Other Countries" figure here excludes Venezuela, which was reported as its own fifth segment that year, for a like-for-like comparison. **FY2017's total direct-contribution margin of 21.94% includes Venezuela's segment, which is no longer part of the 2018 total.

Brazil's direct contribution margin nearly halved for the full year - to 11.95% from 31.73% - confirming that Q2's collapse to 6.85% and Q3's 9.43% were a genuine full-year pattern, not a temporary dip, even as the segment's own revenue grew a healthy 25.4%. Argentina held up meaningfully better across the full year (32.40%, -7.54pp) than its currency turmoil alone would suggest, though its revenue growth (+4.8%) was the slowest of any segment - almost entirely a function of the peso's own devaluation eating into dollar-denominated reported revenue while local-currency GMV grew much faster. Mexico posted the sharpest full-year improvement this coverage has measured (+58.32pp) on 112.5% revenue growth, though it remains the only segment with a negative direct contribution margin. Brazil and Argentina together are still 86.4% of consolidated revenue, so their combined trajectory - Brazil's structural margin erosion against Argentina's relative resilience despite a worse currency environment - remains what drives the consolidated result.

Beyond the Usual

The buyer protection program's maximum exposure grew to $988.7 million - a widening, not narrowing, gap against its allowance

As of December 31, 2018, the buyer protection program's disclosed maximum potential exposure was $988.7 million, up from $925.7 million at year-end 2017 and from $633.7 million as of the September 30, 2018 quarter-end (the year-end figure reflects the full year's payment volume rather than a single quarter's, so the two aren't directly comparable quarter-to-quarter). The allowance, by contrast, grew to just $4.15 million from $1.09 million - a roughly 238x gap between disclosed maximum exposure and actual reserve at year-end.

The Company's own view - stated consistently across every filing this coverage has reviewed - is that maximum potential exposure isn't representative of actual expected losses, and historical loss rates support that framing. But a nearly 240x gap between the two disclosed numbers, growing rather than shrinking as the program scales, is still worth tracking each period for whether the allowance methodology is keeping genuine pace with real claim experience.

Retiring $263.7 million of in-the-money convertible debt cost $348.1 million in cash plus 1,044,298 shares

In connection with issuing the new 2028 Notes in August 2018, MercadoLibre repurchased or exchanged and retired $263.7 million principal amount of its existing 2.25% Notes due 2019 (original conversion price: $126.02 per share, versus a stock price that closed 2018 at $292.85 - more than double). The consideration paid for that $263.7 million of principal was $348.1 million in cash plus 1,044,298 shares of common stock - a real premium reflecting how deep in the money the old notes had become. Separately, MercadoLibre unwound a corresponding portion of its existing capped call hedges tied to those notes, receiving back $121.7 million (August) and $14.4 million (November) from the counterparty financial institutions.

This isn't a governance or disclosure problem - it's simply what it costs to retire convertible debt once the underlying stock has appreciated well past the conversion price - but it's a real, quantifiable cost of MercadoLibre's own share-price appreciation that doesn't appear anywhere as a single line item, only reconstructable from the debt footnote's narrative description.

A $284.3 million Central Bank of Brazil mandatory guarantee is now embedded in the "short-term investments" line

Of the $461.5 million of short-term investments reported on the balance sheet at December 31, 2018, $284.3 million is held as a mandatory guarantee with the Central Bank of Brazil - a regulatory capital requirement tied to MercadoPago's operations as a payments institution in Brazil, disclosed separately from the Company's otherwise-liquid investment portfolio. Unlike the rest of the Company's short-term investments, this amount is not available to be readily converted to cash for general corporate purposes.

A reader comparing MercadoLibre's $461.5 million short-term investments balance against prior years without this detail would overstate the Company's actually-available liquidity by more than 60% of that line item - a genuinely useful footnote-level distinction that doesn't surface anywhere in the summary balance sheet itself.

Executives forfeited the entire 2018 Long Term Retention Plan tranche after its performance targets went unmet

MercadoLibre's Long Term Retention Plan compensation footnote discloses that "because the targets planned were not achieved in 2018, executive officers will not receive any compensation of the 2018 Plan" - the first time this coverage has seen a full forfeiture of an LTRP tranche rather than a reduced payout tied to a falling stock price. Total LTRP compensation expense across all outstanding tranches (2009 through 2018) fell to $27.5 million for the year, down 22.9% from $35.7 million in 2017.

This is a genuinely different mechanism than the stock-price-linked LTRP expense swings this coverage has tracked in prior quarters - a full forfeiture implies the 2018 plan was tied to a specific operating or financial performance target (rather than purely the stock price), and that target wasn't hit, consistent with the year's operating losses and margin compression described above.

Coverage Table

Theme	FY2018	FY2017	YoY	Why it matters
Net revenue	$1,439.7M	$1,216.5M	✅ +18.3%	GMV grew only 6.4% - the widest gross-vs-net gap this coverage has measured, mostly currency
Net (loss) income	$(36.6)M	$13.8M	🔴 First full-year loss this coverage has recorded	Entirely a first-half phenomenon - Q4 alone was near-breakeven
Q4 2018 standalone net loss (derived)	$(2.3)M	n/a	⚠️	Sharp sequential improvement from Q2's $(11.25)M and Q3's $(10.1)M standalone losses
Gross margin	48.4%	59.2%	🔴 -10.8pp	Free-shipping subsidies nearly tripled to $424.8M
Brazil direct contribution margin	11.95%	31.73%	🔴 -19.78pp	A full-year confirmation of the collapse first seen in Q2
New capital raised	$880M (2028 Notes, August 2018)	n/a	New this year	Retiring the old 2019 Notes cost $348.1M cash plus shares for $263.7M of principal

Stock Price - A Round Trip That Ended Roughly Where It Started

MELI shares closed 2018 at $292.85, essentially flat with where the stock traded through much of 2017's second half, but the path there was volatile: a February 2018 peak of $387.97, a May 2018 trough of $290.83 (a 25.0% peak-to-trough decline), a partial recovery to $351.97 by November, and then a 16.8% pullback in December alone to close the year at $292.85. Over the trailing two years (against a January 2017 starting point, the earliest comparable month in this window), the stock is up roughly 58.0%. MercadoLibre has never split its common stock since its August 2007 IPO, so no split adjustment applies to any of these figures.

A year that ends essentially flat, after a round trip through a 25% drawdown and partial recovery, is a genuinely different story than either endpoint alone would suggest - the market spent all of 2018 repricing MercadoLibre's growth-for-margin trade in real time, methodically, rather than settling on a single verdict.

Target Valuation Range

DCF fair-value range: roughly $5,780 million (base case) to $11,300 million (bull case) enterprise value, against an actual $13,508.1 million enterprise value - overvalued. The base-case DCF covers well under half of enterprise value and even the bull case falls short, though the gap has narrowed slightly from Q3 as the stock itself gave back its Q3 gains by year-end.

Market cap → enterprise value	Q4 2018
Share price (period-end, December 31, 2018)	$292.85
Shares outstanding	45,202,859
Market capitalization	$13,237.7 million
Less: cash, cash equivalents & restricted cash	$464.7 million
Plus: loans payable & other financial liabilities	$735.2 million
Enterprise value	$13,508.1 million

Short and long-term investments of $737.7 million - which include the $284.3 million Brazilian Central Bank guarantee described above - are excluded here for simplicity, consistent with prior quarters' methodology. FY2018's actual net revenue of $1,439.7 million is used as a clean trailing-twelve-month figure (no blending or restatement adjustment needed this time, since this is a full fiscal year).

Peer-multiple sanity check	Q3 2018	Q4 2018 (FY2018)	Change
TTM Net revenue	$1,448.6 million	$1,439.7 million	⬇ down
Enterprise value	$15,008.4 million	$13,508.1 million	⬇ down
EV/Sales	10.36x	9.38x	⬇ down

Enterprise value fell more than revenue did with the stock's own year-end pullback. Net income remains negative for the year, so a P/E multiple isn't meaningful.

DCF (base/bull), illustrative: both scenarios start from FY2018's actual free cash flow of $133.2 million (a 9.2% margin on $1,439.7 million of revenue).

Scenario	Key assumption	Implied EV	% of actual EV
Current (Q4 2018 close)	actual market price	$13,508.1 million	100%
Base	Revenue growth decelerating 19%→11% over 5yrs; FCF margin recovering to 21% by year five as shipping-subsidy/marketing spend normalizes; 12% discount, 4% terminal growth	$5,780 million	42.8%
Bull	Revenue growth 23/19/17/14/12%; FCF margin recovering faster, 14%→26% by year five (growth push pays off in durable share gains); 10.5% discount, 5% terminal growth	$11,300 million	83.6%

The base-case percentage is essentially unchanged from Q3's ~42.7%; the bull case's 83.6% is slightly closer to clearing today's price than Q3's ~85.2% miss in relative terms, since Q3's own EV was even further from being cleared.

Reverse DCF: holding a 12% discount rate and solving for the perpetual FCF growth rate that would justify today's $13,508.1 million enterprise value on the $133.2 million FCF base implies a required perpetual growth rate of approximately 10.9%, forever - essentially unchanged from Q3's ~10.8%.

The question that will resolve this valuation gap is the same one flagged in the opening section above: whether Q4's near-breakeven result is the start of a genuine margin recovery toward the base case's assumptions, or a seasonal blip inside a business still fundamentally trading growth for profitability. The next quarter's filing is where that question starts to get a real answer.

---