APH · Amphenol Corporation · Stock research

Current Thesis

The binary flagged in the last refresh resolved in favour of the bulls. On 2026-07-29 Amphenol reported Q2 sales of $8.758B against an $8.239B consensus (+55% in USD, +30% organic) and adjusted diluted EPS of $1.35 versus $1.18 expected, up 67% YoY. The number that mattered most for the interconnect narrative — the order book — printed $10.7B for a book-to-bill of 1.23:1, holding the peak-cycle reading set in Q1 ($9.4B orders, 1.23–1.24 area) rather than rolling toward 1.0. Management guided Q3 to $9.3–9.4B in sales (+50–52%) and $1.40–1.42 adjusted EPS (+51–53%), both well clear of the $8.683B / $1.27 consensus, and raised the CommScope CCS contribution for 2026 to $4.6B of sales and $0.30 of EPS accretion from $4.1B / $0.15. Four desks lifted targets the next morning (UBS $197, Citigroup $210, BNP Paribas $215, Truist $215). Price closed 2026-08-07 at $169.18, 4.0% under the $176.32 52-week high, with RSI(14) at 67.7 after a 32.3% three-month run. The AI-interconnect leg is intact and now twice-ratified by the sell-side; what has changed is that the next company-specific resolution is roughly eleven weeks away.

Bullish and bearish views on Amphenol Corporation

The model's bull view on Amphenol Corporation (APH), in brief: Q2 2026 (2026-07-29): sales $8.758B vs $8.239B consensus; adjusted EPS $1.35 vs $1.18; GAAP diluted EPS $1.37, +59% YoY. The bear view: Organic growth decelerated: +33% in Q1 to +30% in Q2; reported growth eased from +58% to +55%. Both cases follow in full.

Bull Case

• Q2 2026 (2026-07-29): sales $8.758B vs $8.239B consensus; adjusted EPS $1.35 vs $1.18; GAAP diluted EPS $1.37, +59% YoY. Beat on both lines, not a mix-driven EPS beat.
• Orders $10.7B, book-to-bill 1.23:1 (2026-07-29). Orders rose from $9.4B in Q1 to $10.7B in Q2 — the forward pipe expanded sequentially while shipments were already at a record.
• Q3 guidance $9.3–9.4B sales and $1.40–1.42 adjusted EPS (2026-07-29) sits above the $8.683B / $1.27 consensus on both lines; the raise, not just the beat, drove the 2026-07-30 target cluster.
• CommScope CCS 2026 contribution raised to $4.6B sales and $0.30 EPS accretion from $4.1B / $0.15 (2026-07-29) — the $10.5B January acquisition is running ahead of its own underwriting.
• Communications Solutions grew 85% YoY with 42% organic growth in Q2 (2026-07-29), the segment carrying IT datacom and the AI cluster interconnect content.
• Adjusted operating margin 29.8% and free cash flow $1.2B in Q2 (2026-07-29) — the volume ramp is converting to cash, not just to receivables.

Bear Case

• Organic growth decelerated: +33% in Q1 to +30% in Q2; reported growth eased from +58% to +55%. The second derivative is flat-to-down while the multiple prices acceleration.
• Book-to-bill 1.23:1 is a plateau against Q1's 1.24. A ratio in this band is a peak-cycle reading, and hyperscaler capex is an input Amphenol rides rather than steers.
• Sell-side has caught up again. Per S&P Global data carried by CNN Markets as of 2026-08-09, 17 analysts average a $198.06 target with a $170 low and a $230 high — the low end sits essentially at the 2026-08-07 close of $169.18.
• Headline growth remains acquisition-flattered. With CommScope CCS guided to $4.6B of 2026 sales plus El.Com and Wilder Technologies closed in Q2, reported-versus-organic spread stays wide, and integration slippage would surface in Communications Solutions first.
• Policy overhang with no date attached: Reuters reported on 2026-08-04 that the Trump administration is drafting a ban on Chinese data-center devices. The direction of the effect on a US connector supplier with large China manufacturing is genuinely ambiguous, and the draft has no published effective date.
• The optical/interconnect complex trades as one block: on 2026-07-28 Corning's report knocked Coherent, Lumentum and Ciena lower. A single peer's guide-down can compress the multiple here regardless of Amphenol's own order book.

Setup & Price Structure

The 2026-08-07 close of $169.18 sits 4.0% below the $176.32 52-week high. That high postdates the prior note (the highest confirmed mark through 2026-07-18 was the June 22 print of $168.75), so it was set in the post-earnings window — an inference from the price basis, not a quoted intraday tape. Structure below spot: the January 27 all-time high of $167.04 and the June 22 high of $168.75 form the first shelf, now inches beneath price; the June launch shelf and rising 20-week EMA sit near $148–152; the deeper base floor is the $143–145 June consolidation.

Life-cycle: MATURING. The dating is specific — two separate target-raise clusters inside three weeks (six desks 2026-07-09 to 2026-07-13 at $175–200; four desks 2026-07-30 at $197–215), a beat-and-raise that was rewarded with a high the stock has since given back 4.0% of, and a two-for-one split announced 2026-08-05. Well known, still working, moderating flow.

Crowding and positioning observables, stated as observables: RSI(14) 67.7 on 2026-08-07 following a +32.3% three-month return, with price failing to hold the high; retail-facing coverage clustering visible in the Benzinga "$1,000 invested N years ago" series on 2026-06-08, 2026-06-18 and 2026-08-05; a split whose accessibility effect is an attention event by construction; consensus mean target $198.06 roughly 17% above the 2026-08-07 close, leaving nominal re-rate room but with the low target at spot. No insider transactions appear in the 30-day filing record, and there is no earnings date inside the next 30 days — the near-term calendar is mechanical.

Catalyst Calendar (next 30 days)

• 2026-08-17 — Record date for the two-for-one split. Holders of record receive one additional share per share held.
• 2026-09-02 — Additional split shares distributed. Quoted price and every historical chart level rebase by half on the ex-date that follows; a $152 pre-split level reads $76 afterwards.
• 2026-09-22 — Record date for the Q3 dividend of $0.25 per share ($0.125 post-split), payable 2026-10-14. Just outside the 30-day window at the time of writing.
• ~2026-10-28 (est.) — Q3 2026 print. Unconfirmed; Q2 landed 2026-07-29. This is the next reading on orders and book-to-bill, and the next test of the $9.3–9.4B guide.

Elapsed catalysts

• Undated — Finalisation of the draft US rule on Chinese data-center devices reported by Reuters on 2026-08-04. No published timetable. (passed 5d ago)

What Would Change Our Mind

The order book is the load-bearing element. Book-to-bill printing under 1.1 at the Q3 report, or a Q4 guide below consensus after two quarters of guiding above it, would say the AI-interconnect demand curve has flattened — and at a mid-cycle multiple that de-rates quickly. Organic growth is the second gauge: a print below the mid-20s while reported growth stays above 40% would show the acceleration is now acquisition arithmetic.

On price, a weekly close below $152 loses the June breakout shelf and the rising 20-week EMA and reverts the whole post-June structure to base. Because the split distributes 2026-09-02, that same condition reads $76 on a split-adjusted basis thereafter; the level, not the ticker's quoted number, is what is being watched.

A third condition is participation. If the shelf at $167–169 gives way and the stock fails to reclaim $176.32 while SMH and the broader AI-infrastructure complex make new highs, the theme has flipped to SATURATED for this name specifically — the split and the target raises would have been distribution events rather than a fresh bid.

Correlation Notes

• Amphenol is a duplicate exposure to the AI-datacenter basket: SMH, NVDA, ANET, TEL and the optical names all discount the same hyperscaler capex line. The 2026-07-28 Corning-driven selloff in Coherent, Lumentum and Ciena is the recent demonstration that a single supplier's guide moves the whole complex.
• Post-CommScope CCS, the name also carries fibre/broadband cyclicality that the pure connector franchise did not, tying part of the revenue base to carrier and MSO capex rather than to GPU cluster builds.
• The draft US restriction on Chinese data-center devices (Reuters, 2026-08-04) is a sector-level variable that would hit peers and Amphenol through the same channel; it is not an idiosyncratic input.
• Charting versus peers breaks at the 2026-09-02 distribution unless the comparison series is split-adjusted.