Arrow Electronics, Inc. (NYSE: ARW) - Deep Dive Research Report

Prepared 2026-08-16. Most recent reporting period: Q2 2026 (quarter ended ~July 4, 2026), released August 6, 2026.

Section 1: What the company does

Arrow Electronics is a middleman. It sits between the companies that make electronic parts and enterprise IT hardware, and the companies that build things out of them. It buys semiconductors, capacitors, connectors, sensors, power supplies, servers, storage arrays, security software and cloud subscriptions from thousands of manufacturers, holds a lot of it in inventory, and resells it - in the right quantity, to the right factory or reseller, at the right time, with a layer of engineering and logistics wrapped around it. In 2025 it moved roughly $31 billion of this through its books.

That description sounds unglamorous, and the gross economics are thin: Arrow operates on low-single-digit operating margins. But the reason a $31-billion-revenue business exists in the gap between, say, Texas Instruments and a Bosch factory is that neither side wants to deal with the other's complexity. A single industrial or automotive product might use 400 different components from 40 different suppliers. No manufacturer wants 40 supplier relationships, 40 sets of minimum order quantities, 40 lead-time negotiations, and the working-capital hit of holding all that inventory. And no chipmaker wants to invoice, credit-check, warehouse for and ship to 100,000 small and mid-sized customers. Arrow aggregates both sides: it is the single throat to choke for the buyer and the outsourced sales-and-logistics arm for the supplier.

Arrow runs two very different businesses under one roof:

• Global Components (~70% of sales): distributing electronic components - semiconductors, passives, interconnect, electromechanical, power - to original equipment manufacturers (OEMs) and electronic manufacturing services (EMS) firms.
• Global Enterprise Computing Solutions / ECS (~30% of sales): distributing datacenter and enterprise IT - servers, storage, software, cloud, cybersecurity - to value-added resellers (VARs) and managed service providers (MSPs).

The company traces back to 1935, when it started as a retail radio-parts store in New York City ("Arrow Radio"). Over ninety years it transformed from a parts shop into one of the two largest global electronics distributors, largely through an acquisition-heavy strategy through the 1980s-2010s (Schweber, Wyle, Richardson RFPD, immixGroup, eInfochips and dozens more) that bolted on product lines, geographies and, latterly, engineering and software capability. Its tagline, "Five Years Out," reflects a positioning it has pushed for a decade: that Arrow is not just moving boxes but helping customers design tomorrow's products.

The essential thing to understand about Arrow is that it is a spread business on a cyclical commodity. When the components cycle is tight (shortages, long lead times, price inflation), Arrow's revenue, margins and inventory profits all inflate together. When the cycle turns (excess inventory, cancellations, price deflation), all three deflate together. The six quarters covered in this report capture Arrow climbing out of the deepest components downturn since 2019 and back into a sharp recovery - revenue went from declining in early 2025 to growing 39% year-over-year by Q1 2026.

Rick Marano, President of Global Components, on the Q2 2026 call framed the recovery as bifurcated: the AI-driven datacenter buildout and a separate, still-early cyclical recovery in traditional industrial, automotive and aerospace end-markets are both lifting the business at once.

Section 2: Business segments

Global Components (~70% of consolidated sales)

What it does. This is the core historical Arrow: authorized distribution of electronic components. Arrow buys from component manufacturers (semiconductor makers like Analog Devices, Texas Instruments, Infineon, STMicroelectronics, NXP, onsemi; passive and interconnect makers; power and electromechanical suppliers) and sells to the companies that assemble finished electronics. Its customers are OEMs designing and building industrial equipment, automotive electronics, aerospace and defense systems, medical devices, transportation and telecom gear, plus the EMS contract manufacturers who assemble on their behalf. In Q2 2026 the segment did roughly $7.4 billion of sales, with strength called out specifically in aerospace/defense, industrial and transportation verticals, and book-to-bill above 1.0 in all three regions (Americas, EMEA, Asia-Pacific).

Core capability. Three things that took decades and a lot of acquired scale to build. First, breadth of line card - Arrow carries one of the widest supplier portfolios in the industry, so a customer can source most of a bill-of-materials from one distributor. Second, design-chain engineering - Arrow employs field application engineers who help customers select components and design them into new products at the earliest stage. A component "designed in" at the prototype stage generates years of downstream volume, and it is very sticky. Third, supply-chain services - inventory management, kitting, programming, forecasting and demand-planning that let a customer outsource the working-capital and logistics burden. The engineering layer is what separates Arrow (and Avnet) from pure catalogue distributors.

Why it's separate. Components distribution has fundamentally different economics, suppliers, customers and working-capital dynamics from IT distribution. It is inventory-heavy, engineering-led, and tied to the semiconductor cycle. It is a genuinely different business from reselling servers to IT resellers.

Competitive position. The direct global peer is Avnet. In Asia, WT Microelectronics (which acquired Future Electronics in 2024) and WPG Holdings are very large. In the high-service, small-order, catalogue/online niche, Digi-Key and Mouser compete on the long tail. Arrow wins on breadth of line card, design-engineering depth and global supply-chain scale; it is more exposed on the pure-catalogue, fast-turn small-order segment where Digi-Key and Mouser are stronger.

Fit in the group. This is the engine - the largest, most cyclical, and highest structural-margin segment (Global Components operating margin reached 5.4% in Q2 2026 vs low-single-digits for ECS). When management talks about operating leverage, this is where it comes from: fixed cost base, so incremental recovery revenue drops through at high rates.

Global Enterprise Computing Solutions / ECS (~30% of consolidated sales)

What it does. ECS distributes datacenter and enterprise IT - servers, storage, networking, infrastructure software, cybersecurity, cloud and analytics - from suppliers like HPE, IBM, Dell, NetApp, Microsoft, VMware/Broadcom and others, to a channel of value-added resellers and managed service providers who in turn serve end enterprises. Arrow adds credit, aggregation, technical enablement, cloud-marketplace tooling (its ArrowSphere platform) and increasingly takes over pieces of suppliers' own go-to-market via multiyear strategic-outsourcing contracts. Q2 2026 ECS sales were ~$2.6 billion, up 14% year-over-year.

Core capability. Channel enablement and recurring/subscription orchestration. ECS knows how to help thousands of small resellers sell, provision and bill complex hybrid-cloud and software subscriptions, and to manage the credit and logistics behind that. ArrowSphere - its cloud/subscription management platform - is the piece management points to as differentiated and recurring.

Why it's separate. Completely different suppliers, customers (resellers not manufacturers), sales motion (software/subscription/services rather than physical components), and margin structure. It is a mostly asset-light, billings-based business where reported revenue can be recognized net (agent) for some software/cloud, which distorts headline revenue vs the underlying billings.

Competitive position. Here Arrow competes with the IT-distribution giants: TD Synnex and Ingram Micro globally, plus regional players. It is a scale-and-relationships business with thinner differentiation than Components; the recent multiyear strategic-outsourcing contracts are Arrow's attempt to move up the value chain, but they have caused near-term pain (see below).

Fit in the group. The diversifier and the recurring-revenue option. It smooths the components cycle (IT spend is less violently cyclical than the chip cycle) and offers a subscription/cloud growth angle. But in the last three quarters it has been a source of negative surprises: a $21 million charge in Q3 2025 and a $27 million charge in Q2 2026, both tied to underperforming multiyear contracts with a partner where Arrow took on non-core parts of a supplier's business and mis-estimated the profitability. Management insists these contracts are accretive long-term but is clearly still working through "growing pains."

Section 3: Products and business detail

The Components catalogue. Arrow's Global Components line card spans the full electronic bill-of-materials: semiconductors (microcontrollers, analog and power ICs, memory, FPGAs, discretes, sensors, RF), passive components (resistors, capacitors, inductors), interconnect (connectors, cable), electromechanical (relays, switches, thermal), power products (supplies, converters, batteries), and embedded computing boards. Within this, Arrow's Richardson RFPD unit is a specialist in RF, wireless, power and energy components - a higher-engineering-content niche. The value here is not the parts themselves (anyone can buy a resistor) but the authorized relationship: Arrow is a franchised distributor, meaning the manufacturer warrants the parts through Arrow's channel, which matters enormously given that the ECIA estimates up to 5% of components in the open market are counterfeit. Buying through an authorized distributor is a traceability and warranty guarantee.

Value-added services. This is where Arrow earns above bare distribution margin. Services include: design-chain engineering (field application engineers embedded with customers at the design stage); programming and kitting (pre-loading firmware onto microcontrollers, assembling ready-to-use kits); supply-chain management (vendor-managed inventory, demand forecasting, buffer-stock programs); and, through acquired capabilities like eInfochips, actual product-engineering and design services (turnkey product development, embedded software, digital engineering). eInfochips lets Arrow move upstream into designing the customer's product, not just supplying parts for it.

ArrowSphere and ECS tooling. On the IT side, ArrowSphere is Arrow's cloud-and-subscription management platform - a marketplace through which resellers procure, provision, manage and bill multi-vendor cloud and software subscriptions. It is the recurring, platform-economics piece of ECS that management highlights as strategically important.

Manufacturing and delivery. Arrow manufactures nothing. Its "factory" is its logistics and engineering network: over 140 sales facilities and roughly 39 distribution and value-added centers, serving more than 85 countries with over 460 total locations and ~17,000 employees. The physical process is warehousing, kitting, programming and shipping; the intellectual process is the engineering and demand-planning wrapped around it. The binding constraint on the business is working capital and inventory positioning - holding enough of the right parts to serve demand without being caught long when the cycle turns.

Geographies. Sales split across three regions - the Americas, EMEA, and Asia-Pacific. Asia-Pacific was the first region to show the components recovery (called out in Q2 2025), consistent with Asia being closest to the manufacturing base. Foreign-currency movements are a recurring swing factor on reported growth (a ~$84 million FX drag on Q1 2025 sales, for example).

Milestones that shaped the current business. From a 1935 radio-parts store, the pivotal moves were: the acquisition-driven scaling into the #1/#2 global components distributor through the 1980s-2000s; the build-out of ECS as a second leg through IT-distribution acquisitions (including immixGroup in public-sector IT); the push into engineering services via eInfochips (2018); and the current transformation - repositioning from box-mover to "strategic partner," including taking on multiyear strategic-outsourcing contracts in ECS where Arrow runs non-core functions for suppliers.

Section 4: Customers

Who buys. Two entirely different customer bases. On the Components side: over 100,000 OEM and EMS customers - the companies that design and build industrial equipment, cars, aircraft, medical devices, telecom and networking gear. These range from global blue-chips to mid-sized manufacturers. On the ECS side: value-added resellers and managed service providers - the IT channel that in turn serves enterprises and public-sector buyers.

Who makes the decision, and on what criteria. In Components, the buying decision is split: the design engineer chooses which parts get "designed in" (driven by technical fit and the quality of Arrow's application-engineering support), while procurement chooses the sourcing channel (driven by availability, lead time, price, credit terms and inventory service). The design-in decision is the strategic one - it happens early, has a long sales cycle (a new industrial or automotive platform can take a year or more to design and then produces for years), and it locks in downstream component volume. In ECS, the reseller chooses a distributor based on vendor breadth, credit lines, technical enablement, cloud tooling and rebate economics.

Why they choose Arrow. Breadth (source most of a BOM from one place), authorized/traceable supply (counterfeit avoidance and warranty), engineering support at design stage, supply-chain services that offload working capital, and global consistency for multinational customers who want the same distributor in every region.

Switching costs. Moderate-to-high on the Components side once a design-in has happened: re-sourcing a qualified component through a different distributor is friction, and the embedded field-engineering relationship is sticky. But at the procurement/spot level, distribution is competitive and price-sensitive - a customer can and will multi-source across Arrow, Avnet, Digi-Key and others for availability. On the ECS side, switching costs are lower and driven by vendor authorizations and credit terms.

Concentration. Low customer concentration - with 100,000+ customers, no single buyer dominates, which is a quality feature (diversified demand). Supplier concentration is the more relevant dynamic: Arrow depends on a relatively small number of large semiconductor and IT vendors for its franchise line card. Losing an authorization from a major supplier, or a supplier taking distribution direct, would matter far more than losing a customer. The recent Analog Devices decision to name Arrow a strategic global distribution partner illustrates how consequential a single supplier relationship can be.

Contract structure and revenue predictability. Mostly transactional, book-and-ship distribution rather than long-dated contracts - which is why book-to-bill and backlog are the metrics management watches. In the recovery, book-to-bill has run above parity across all three regions with backlog building into 2027, giving unusually good near-term visibility. The exception is ECS's multiyear strategic-outsourcing contracts, which are long-dated - and which have proven to be the source of the recent margin charges when the economics ran below plan.

Section 5: Competitive landscape

Electronics distribution is a scale game with two distinct arenas. In components distribution, the structure is an oligopoly at the top (Arrow and Avnet globally; WT Microelectronics and WPG dominant in Asia) plus a strong high-service catalogue tier (Digi-Key, Mouser, Newark/Farnell) for the small-order long tail. In IT/enterprise distribution, Arrow's ECS competes against the two IT-distribution giants, TD Synnex and Ingram Micro.

Where Arrow wins. Against Avnet, Arrow has generally run a bit larger and has pushed harder into higher-value engineering (eInfochips) and cloud (ArrowSphere). Its global line-card breadth and design-engineering depth are genuine advantages over pure catalogue players. Against the Asian distributors, Arrow's edge is in the Americas and EMEA and in design-led Western OEMs.

Where Arrow is exposed. In the fast-turn, small-order, online catalogue niche it is weaker than Digi-Key and Mouser, who have built superior e-commerce and inventory-availability models for that segment. In Asia, the WT-Future combination created a formidable scale competitor. And in ECS, Arrow is a sub-scale #3 behind TD Synnex and Ingram Micro. The whole industry is structurally low-margin and price-competitive - this is a spread business, not a moat business. Arrow's defensibility comes from scale, supplier authorizations and engineering stickiness, not from pricing power.

Barriers to entry. High for a new global full-line distributor (you would need supplier authorizations, global logistics, working capital in the billions, and design-engineering headcount that takes years to build), but low for niche entrants and near-zero switching costs at the spot-procurement level. The bigger structural threat is disintermediation: large suppliers or large customers deciding to transact direct, or online marketplaces compressing the catalogue tier.

Market caps are approximate, move daily, and are shown only as a peer-size reference as of August 2026.

Section 6: Industry

Demand drivers. Components distribution demand ultimately tracks electronic content in the physical economy: industrial automation, automotive electrification and ADAS, aerospace and defense build rates, medical devices, telecom infrastructure, and the general secular rise in semiconductor content per product. ECS demand tracks enterprise IT spending, datacenter buildout, hybrid-cloud adoption, cybersecurity, and - increasingly - AI infrastructure. Rick Marano's Q2 2026 framing is the key industry insight: two demand waves are running at once, an AI/datacenter wave and a separate, still-early cyclical recovery in traditional end-markets.

Size and growth. Estimates for the global electronic-component distribution market vary widely by scope and methodology, clustering in the ~US$180-200 billion range for 2024-2025 on the narrower definitions (one source has it advancing from ~US$186bn in 2024 to ~US$199bn in 2025, ~7% CAGR to the early 2030s), with much larger figures on broader definitions that include IT and end-distribution. Semiconductors are the majority of the pie (roughly half or more of distributed value). Arrow, at ~US$31bn of 2025 sales across both segments, is one of the two largest players in the world.

Position in the supply chain. Arrow is the channel layer between component/IT manufacturers and the OEM/reseller base - the aggregation, inventory, credit and engineering node. It captures a spread and adds service value; it does not own the IP or the fabs.

Regulation and certification. The critical regulatory/quality dynamic is counterfeit avoidance - authorized-distribution status (franchise agreements, traceability, warranty) is the industry's answer to a counterfeit rate the ECIA puts at up to 5% of the open market. Trade policy and export controls (semiconductor export restrictions, tariffs, country-of-origin rules) are a growing overlay, particularly for the Asia-Pacific business. Price inflation from tariffs was even called out as a contributor to recent revenue growth.

Cyclicality. This is the defining industry characteristic. Components distribution is one of the most cyclical corners of tech: shortages drive lead-time extension, price inflation, double-ordering and inventory-building (inflating distributor revenue and margins), then the cycle inverts into cancellations, excess inventory, price deflation and destocking. Arrow's last three years are a textbook cycle - the 2023-2024 downturn (revenue declining through early 2025) followed by the sharp 2025-2026 recovery (revenue +39% YoY by Q1 2026). ECS is meaningfully less cyclical, which is part of its strategic rationale.

Tailwinds: AI datacenter buildout, electrification, defense spending, rising semiconductor content, and a cyclical restock underway. Headwinds: structural margin thinness, disintermediation risk, tariff/export-control complexity, and the ever-present risk that today's restock becomes tomorrow's glut.

Section 7: Growth triggers

All items sourced from the six most recent earnings calls (Q1 2025 through Q2 2026).

• Cyclical recovery in Global Components, described as "early stage." Management has repeated across multiple calls that the components upcycle is in its early innings, implying a longer runway of sequential growth. First flagged as a return to YoY growth in Q2 2025 (Aug 2025, first YoY components growth since Q4 2022), reinforced in Q3 2025 (Nov 2025) and again in Q2 2026 (Aug 6, 2026).

  "We are in the early stages of a gradual recovery" in Global Components. (Interim CEO Bill Austen, Q3 2025 call, Nov 2025; echoed by Rick Marano, Q2 2026 call, Aug 6, 2026.)

• Book-to-bill above parity in all three regions with backlog building into 2027. Repeated trigger giving forward visibility. (Q4 2025 call, Feb 5 2026; Q2 2026 call, Aug 6 2026.)

  Book-to-bill "remains well above parity, with backlog building into 2027." (Q2 2026 call, Aug 6, 2026.)

• Two simultaneous demand waves - AI/datacenter and a separate traditional cyclical recovery. Management distinguishes the AI buildout from the broader industrial/auto/aerospace restock, arguing both are lifting the business. (Q2 2026 call, Aug 6, 2026.)

• Customers restoring buffer inventory, "disciplined rather than excessive." Restocking after the destocking cycle is a near-term volume tailwind that management frames as healthy, not a bubble. (Q2 2026 call, Aug 6, 2026.)

• ECS multiyear strategic-outsourcing contracts expected to become margin-accretive. Arrow is taking over non-core functions from suppliers; management says these are dilutive now (hence the Q3 2025 and Q2 2026 charges) but accretive over the contract life. (Q3 2025 call, Nov 2025; Q2 2026 call, Aug 6, 2026.)

• Operating leverage on the fixed cost base as recovery revenue scales. Management repeatedly ties margin expansion (Components op margin to 5.4% in Q2 2026) to disciplined expense management plus incremental recovery volume. (Q1 2026 call, May 7 2026; Q2 2026 call, Aug 6 2026.)

• New leadership transition: Dee Merriwether joining as President & COO in early September 2026, alongside the ongoing permanent-CEO search - a potential strategic-reset catalyst. (Q2 2026 call, Aug 6, 2026.)

• New $1 billion share repurchase authorization (effective May 12, 2026). A capital-return catalyst as free cash flow recovers. (Announced May 13, 2026, discussed on the Q1 2026 call.)

Section 8: Key risks

• Cycle reversal / inventory glut. This is the dominant risk. The current revenue surge is partly restock and partly tariff-driven price inflation. If customers over-order into buffer stock and the cycle turns, Arrow faces the classic distributor whipsaw: cancellations, excess inventory written down to falling prices, and revenue and margins deflating together. Management itself is watching this - Marano's insistence that restocking is "disciplined rather than excessive" (Q2 2026) is precisely the reassurance a management team offers when the market is worried about the opposite. Mechanism: high probability of eventual cyclical reversal; the question is timing and severity. This is a structural feature, not a tail risk.

• ECS strategic-outsourcing contracts keep bleeding. Two consecutive charges - $21m (Q3 2025) and $27m (Q2 2026) - on multiyear contracts with "one partner" that Arrow underestimated. If more of these contracts prove mis-priced, the "margin-accretive over time" thesis breaks and ECS becomes a recurring source of negative surprises rather than a diversifier.

  The Q2 2026 charge related to "underperforming multiyear contracts with one partner." (Q2 2026 call, Aug 6, 2026.) Two charges in three quarters on the same theme is a pattern, not a one-off, and it warrants scrutiny of how these contracts were underwritten.

• Leadership vacuum. Arrow has been run by an interim CEO (board member Bill Austen) since Sean Kerins departed abruptly in September 2025, with a permanent-CEO search still open nearly a year later. A new President/COO (Dee Merriwether) arrives September 2026. Extended interim leadership during a cyclical inflection risks strategic drift, and a new permanent CEO could reset strategy, capital allocation or the ECS contract approach. Management has said the Kerins separation was not tied to a strategic shift, but the length of the search is itself a mild concern.

• Supplier disintermediation / authorization loss. Arrow's franchise depends on a concentrated set of large suppliers. A major supplier pulling an authorization, going direct, or shifting to a competitor (as suppliers periodically rationalize their distribution) would hit a large slice of revenue. The flip side - Analog Devices naming Arrow a strategic partner - shows how binary these relationships can be.

• Thin margins amplify small errors. At low-single-digit operating margins, modest gross-margin pressure, an FX swing (an ~$84m FX drag hit Q1 2025 sales), a bad inventory position or a working-capital squeeze translates into an outsized earnings move. There is very little cushion.

• Tariff and export-control exposure. Price inflation from tariffs is currently helping reported revenue, but the same policy environment can reverse (demand destruction from higher prices, or export restrictions cutting off Asia-Pacific component flows). What is a tailwind today is a two-sided risk.

Section 9: Walk the talk

The six calls used for this assessment: Q1 2025 (May 1, 2025), Q2 2025 (Aug 7, 2025), Q3 2025 (Nov 6, 2025), Q4 2025 (Feb 5, 2026), Q1 2026 (May 7, 2026), and Q2 2026 (Aug 6, 2026). The most recent is within ~10 days of this report. Note the mid-window CEO change: Kerins ran the first two calls; Austen (interim) has run the last four.

The through-line of these six calls is a management team that called the recovery early and then consistently beat its own conservative guidance as the cycle inflected in its favor - which is easier to do on the way up than on the way down, and worth remembering.

In Q1 2025 (May 2025), Kerins struck a cautious tone: revenue was still declining 2% year-over-year (flat constant-currency), and the message was that results "exceeded our guidance ranges" against a soft backdrop. The guide-low, beat pattern was already in place - GAAP EPS came in at $1.51 against modest expectations.

By Q2 2025 (Aug 2025), the pivotal call, Kerins delivered the first concrete inflection: Global Components grew year-over-year for the first time since Q4 2022, led by Asia and improving industrial/transportation trends. He also flagged that lead times had stabilized at pre-pandemic levels - a claim that proved durable through subsequent quarters. This was the promise that mattered, and it was kept: the recovery Kerins pointed to did materialize and accelerate.

Q3 2025 (Nov 2025) was the first Austen call after Kerins' abrupt September departure. Two things stand out. Management reiterated the "early stages of a gradual recovery" framing (delivered on) - revenue grew 13% YoY. But it also dropped the first ECS charge ($21m) on underperforming outsourcing contracts, which had not been flagged in prior calls. This is the clearest "quietly emerged" negative of the window: the strategic-outsourcing contracts were sold as a growth positive, and the downside surfaced only when it hit the P&L.

Q4 2025 (Feb 2026) validated the recovery thesis hard: revenue $8.7bn, +20% YoY, above guidance, EPS $4.39 (+48% YoY), with "steadily rising book-to-bill and expanding backlog across all regions." The Q1 2026 guide (sales $7.95-8.55bn, EPS $2.70-2.90) again looked conservative in hindsight.

Q1 2026 (May 2026) blew past that guide: revenue $9.5bn (+39% YoY) versus a $7.95-8.55bn guide, EPS $5.22 versus $2.70-2.90 - an 85% EPS beat. Either the guide was extremely conservative or the recovery was steeper than management modeled; likely both. The Q2 guide (sales $9.15-9.75bn, EPS $4.32-4.52) once more set a beatable bar.

Q2 2026 (Aug 2026) beat again: $10.0bn revenue (+32% YoY, a record), EPS $5.45 (+124% YoY), above the high end. But the second ECS charge appeared ($27m, up from Q3 2025's $21m, same "one partner" theme), confirming that the outsourcing-contract problem is recurring, not resolved.

Assessment. On the top-line cyclical call, this management (Kerins then Austen) has been credible and, if anything, conservative - it called the components inflection early and repeatedly under-guided into a strengthening cycle. That is a positive credibility signal, with the honest caveat that beating conservative guidance in a recovery is the easy direction. The clear blemish is ECS strategic-outsourcing: sold as a growth positive, it has produced two unflagged charges on the same contracts, which is a legitimate mark against the quality of guidance in that segment. Net read: reliable and slightly sandbagging on the cyclical Components story; less reliable and reactive on the ECS contract economics. The real test of this team's credibility - guiding down gracefully - has not yet come, because the cycle has only gone up during this window.

Section 10: Shareholder friendliness index

Dividends. Arrow Electronics pays no cash dividend and has not during the period reviewed. This is a long-standing policy, not a suspension - Arrow returns capital exclusively through share repurchases. There is therefore no DPS trend to track; the entire capital-return story is buybacks.

Buybacks and dilution. Arrow has an established repurchase program but dialed it down sharply during the downturn to protect the balance sheet, then re-armed as the recovery took hold. Over the last three years: 2023 was heavier (over $300m repurchased, with a large treasury-share retirement of 67.7m shares that year), then activity slowed - roughly 2.0 million shares for ~$250 million in 2024, and only 1.3 million shares for ~$149.9 million in 2025 (leaving ~$172.9m on the prior authorization). In the trailing ~90 days captured by MoatMap's feed, buyback activity was light - a filing dated 2026-08-06 showing consideration lines of roughly $50.7m and $41.9m, and the Q2 2026 call disclosed $43 million repurchased in the quarter - modest against the company's size. The important recent signal is the new $1 billion repurchase authorization effective May 12, 2026 (replacing the prior program), announced as free cash flow recovered - a re-commitment to buybacks as the cycle turns. On dilution: shares outstanding have drifted down modestly over three years (roughly the mid-50-millions range, ~53.8m as of early 2024, trending lower), so the count is shrinking slowly rather than being inflated by option issuance - the buyback has more than offset dilution even at the reduced pace.

Verdict: Returns Capital (buyback-only, cyclically throttled) - Arrow reliably retires shares rather than paying dividends, pulled back prudently in the downturn, and has just re-armed with a fresh $1bn authorization as cash flow recovers.

Section 11: Insider activities

Source: MoatMap cross-market disclosure database (US venue = SEC Form 4 via EDGAR), current as of 2026-08-15, used as the spine and consistent with the open-venue cross-check. Last 12 months: 14 transactions across 10 insiders - 0 open-market buys, 6 open-market sells, 8 "other" (grants / zero-price director-equity awards).

Buys. There were no open-market insider purchases in the last 12 months. No conviction buy signal to read here.

Sells - the why. The selling is concentrated in one person, HR/governance officer Gretchen Zech (5 of the 6 sells), clustered in a tight May-June 2026 window that coincides with the sharp post-recovery run in the stock (sale prices climbing from ~$212 to ~$229). The pattern - an option/award exercise (the "Other" at ~$81) followed same-window by open-market sales at ~$228 - is the classic exercise-and-sell / liquidity-and-diversification behavior of a senior officer monetizing appreciated equity after a strong price move, not a fundamental signal. The individual sells by CLO Carine Jean-Claude and ECS President Eric Nowak are single, modest transactions consistent with routine diversification. None of these is disclosed as tied to a negative view; several are consistent with pre-arranged 10b5-1 plans (common at this seniority), though the plan status is not specified in the block. Where a specific reason is not disclosed, treat it as routine post-appreciation selling rather than a signal.

The May 12 "Other" rows for seven directors at $0 are simply the annual board equity grant (RSUs/deferred stock awarded at the annual meeting) - housekeeping, not a market signal.

Net assessment. Insiders were net sellers, but the activity is (a) narrow (dominated by one HR/governance officer), (b) sourced largely from option/award exercises, and (c) well explained by a stock that roughly doubled off its cyclical lows into the sale window. There is no cluster selling by the CFO or the operating heads that would suggest a shared negative view, and no buying to provide a positive tell. Read: neutral-to-mild-caution. The absence of any insider buying during a powerful earnings recovery is worth noting - insiders monetized rather than added - but the selling itself is routine diversification, not a red flag. (One structural caveat: Arrow is between permanent CEOs, so the highest-conviction insider - a committed CEO - is not present to buy.)

Section 12: Scenarios

Bull case. The two demand waves Marano described both run for years, not quarters. AI datacenter buildout keeps pulling ECS and high-end components, while the traditional industrial-automotive-aerospace restock - genuinely early in mid-2026 - broadens and deepens as global manufacturing re-accelerates. Book-to-bill stays above parity, backlog keeps extending past 2027, and Arrow's fixed cost base delivers powerful operating leverage: each incremental recovery dollar drops through at high margins, and Components operating margin pushes well beyond the 5.4% seen in Q2 2026. Management's habit of under-guiding means the beats keep coming. The ECS outsourcing contracts finally season into the promised accretion, turning a recurring charge into a recurring tailwind and validating the "strategic partner" transformation. A strong permanent CEO arrives, inherits a business at cyclical highs, and deploys the fresh $1bn buyback authorization into a still-modest share count, compounding per-share value. Arrow exits the cycle structurally larger and higher-margin than it entered.

Base case. The recovery is real but behaves like every components cycle eventually does - it moderates. Growth stays positive through 2026 and into 2027 on backlog and restocking, but the year-over-year comparisons cool as the easy recovery laps become hard ones. Components carries the group; ECS grows steadily but keeps absorbing occasional contract lumpiness before the outsourcing deals stabilize. Margins hold in the low-single-digits with the cyclical Components boost, and management continues its guide-low-beat cadence without dramatic upside. The permanent CEO and new COO (Merriwether) settle in without a strategic upheaval, and buybacks resume at a measured pace under the $1bn authorization, shrinking the share count slowly. Nothing breaks; Arrow remains the #1/#2 global distributor delivering steady, cyclical, thin-margin distribution economics - exactly what it has been for decades, at a higher revenue base.

Bear case. Today's surge turns out to be substantially restock-and-tariff-inflation rather than durable end-demand. Customers who built buffer stock in 2026 discover they over-ordered; cancellations and destocking arrive in 2027, and Arrow gets caught long on inventory that has to be marked down against deflating component prices - the classic distributor whipsaw, amplified by thin margins into a sharp earnings decline. The ECS outsourcing contracts prove structurally mis-underwritten rather than merely early, and the charges keep coming, forcing an unwind that dents both segments' credibility. The prolonged leadership vacuum bites: a permanent CEO arrives late, resets strategy mid-downturn, and takes an impairment to clean the slate. A major supplier rationalizes its distribution and pulls or reduces an authorization, lopping off a chunk of Components revenue. Tariffs flip from tailwind to demand-destroyer, and the Asia-Pacific business gets tangled in export controls. The buyback gets throttled again to defend the balance sheet, and Arrow spends 2027-2028 doing what it did in 2023-2024 - grinding through a components glut with little cushion.