CAAP — Corporacion America Airports S.A.

CAAP — Corporación América Airports S.A. · deep-value analysis

28 August 2026 · reference price USD … (NYSE) · 163.178 mn shares outstanding (165,219,146 issued − 2,040,816 treasury, Item 7, 20-F FY2025) · market cap USD 3,946 mn · reporting and trading currency: USD (no currency-mixing risk)

Primary sources: 20-F FY2025 filed 17.03.2026 (SEC-Filings/CAAP/CAAP_20F_2026-03-17.txt), 20-F FY2024/FY2023/FY2022, condensed interim financial statements at 30.06.2026 and the 2Q26 results release (6-K 18.08.2026, accession 0001104659-26-098141 / -098142), the 6-K of 26.06.2026 on Brasília. Data pack: data-pack-CAAP-20260828.md. Research brief: research-CAAP-20260828.md.

Executive summary

The thesis. CAAP operates 52 concessioned airports in six countries, with 86.7 million passengers in 2025, and on the surface it looks exactly like a deep-value opportunity: TTM P/E ~13x, reported net leverage 0.5x EBITDA, an “FCF yield” of 11… on any screener, the first dividend in company history (USD 150 mn approved in August 2026), and coverage of only 7 analysts, with an average target price of USD 32. My thesis is that this profile is an accounting illusion produced by two things screeners can’t see, and that at USD … the stock is fairly-to-slightly-expensively priced, not cheap.

The first thing: CAAP accounts for concessions under IFRIC 12, and investment in concession assets — USD 232 mn over the trailing 12 months — is reported WITHIN operating cash flow, as a separate “Acquisition of Intangible assets” line within operating activities (20-F FY2025, cash flow statement, F-10: −203,165 thousand in 2025; interim 30.06.2026: −107,459 thousand in H1 2026). The “Capital Expenditure” line in investing is only USD 16.7 mn a year. Consequence: anyone doing the reflex CFO − capex gets USD 454 mn and an 11.5% yield, but that’s a flow before interest, before leases, and before the minority share, since the issuer is IFRS and strips interest out of operating. Here the bridge isn’t wrong in the data pack — it’s correct — but it has to be read to the end.

The second thing, more important and completely invisible in any metric: the Brazilian government suspended, in December 2025, the enforceability of the fixed concession fee at Brasília, R…lion (20-F FY2025, the “Concession fees” section for the Brazilian concessions). The amount wasn’t forgiven — it was deferred. It appears in the cash flow statement as a positive addback on the “Unpaid concession fees” line (+55,050 thousand in 2025, +36,003 thousand in H1 2026) and accumulates on the balance sheet as “Concession fee payable”, which rose from USD 748.5 mn (Dec. 2024) to 918.8 mn (Dec. 2025) to USD 1,012.8 mn at 30 June 2026 (Note 23 and interim Note 13). In other words, roughly USD 57 mn a year of CAAP’s “cash flow” is an unpaid bill, not money earned. And, as an order of magnitude: the net debt the company communicates — USD 381.0 mn, 0.5x leverage — ignores a contractual obligation of over a billion dollars, with USD 1,449 mn undiscounted due beyond 5 years (Note 23, maturity table).

The FCF bridge, TTM Jul.2025–Jun.2026: CFO 474.6 mn → simple FCF 454.5 mn → FCFE 338.5 mn → normalized FCFE USD 281.5 mn (7.1% yield on market cap). The gap versus the figure the market publishes (454 mn, 11.5%) is 61% and fully explains the divergence in verdict.

Estimated value. Five triangulated models give a range of USD 13.3 – …/share, with a median of USD …: bear DCF FCFE …, EPV Greenwald with concession debt included …, base DCF FCFE …, own 5-year historical multiples …, bull DCF FCFE (a successful Brasília renegotiation) …. A Monte Carlo simulation over 20,000 scenarios around the base case gives a median of … and a probability of undervaluation of …% — i.e. two chances in three that at USD … you’re paying above value.

Verdict: AVOID at the current price; WATCH below USD 18. Not because the business is bad — the concessions are real, the assets are good, Armenia and Italy are growing nicely — but because today’s price assumes the Brazilian bill will never be paid, that Argentina (55% of EBITDA) doesn’t deteriorate, and that the committed investment program of over a billion dollars over the next five years self-finances without touching the flow to shareholders. Three simultaneous optimistic assumptions, for a 7.1% FCFE yield in a geography with Argentine country risk. The key event that could flip the thesis — the tender for 100% of Inframerica, expected by December 2026 — is four months away and worth waiting for.