Is it a good business? · 6 · What does its record show?
More sales add profit at a steady rate
Sales, $B
16.3
16.0
15.8
17.3
19.9
Sep 2022Sep 2023Sep 2024Sep 2025Sep 2026
Research4.5 to 4.8 centsresearch per $1, steady for five years
Overhead10 to 11 centsselling and office per $1
Cars$7,038M → $7,052Mcar sales, 2023 to 2025
Research and selling costs rise with sales; better factories and price rises widened profit per dollar.
Is it a good business? · 7 · How much room is there to grow?
A $99.2B connector market
About $99.2Bworld connector spending, 2025
+14.7%growth
+31%data-centre power capacity in 2026
Data centres are the biggest buyer and cars second.
Is it a good business? · 8 · Who is in charge?
A long-serving team that shrank the share count
Since 2017Terrence Curtin chief executive
Since 2016Heath Mitts finance chief
Under 1%owned by insiders
The bonus pays on sales, operating profit and earnings per share.
The story right now · 9 · Why do AI and the grid need TE?
AI racks and power lines need more connectors
+34%TE's data-centre unit last quarter
+31%data-centre power capacity in 2026
$6.06Brecord orders
Each AI rack is packed with copper cables and power connectors, and every new data centre needs grid connections. Both are growing far faster than cars.
The story right now · 10 · Will light replace TE's copper?
Light carries data far; copper carries power
Inside a rackShort copper cables.
Between racksLight in glass fibre.
Power to every chipAlways copper.
NVIDIA ships light switchesThe watch point.
The market fears light replaces TE's copper. Light cannot replace a power cable, and TE says copper stays in charge inside the rack.
The story right now · 11 · What changed lately?
Record sales of $5.31B and orders of $6.06B
$5.31Bquarterly sales, up 12%
+34%data-centre unit
$6.06Borders, up 29%
+7%car sales
Strong quarter
It met our test, but the car half has one quarter of growth. The next forecast is for sales up 15%.
The story right now · 12 · What is everyone arguing about?
Can cars grow again beside the AI boom?
Many investors think
A car-parts maker whose AI side may fade as light replaces copper
We think
The AI and grid half keeps growing; cars cap the whole
Analysts split evenly between buy and hold. Against us: data-centre orders falling, or car sales falling while carmakers build more.
The price and our view · 13 · What is a share worth?
The price sits about a tenth below our value
$244what we think a share is worth · the last close is of that
$244our value
Attractive below $195
Neutral
Unattractive above $292
$136gloomy: growth halves and profit per $1 falls
$244our estimate: experts' path, then slowing
$312sunny: growth runs faster
Value per share in three futures. None is a prediction of the share price.
We estimate what a share is worth from the cash the business should earn over many years. Our rule: buy only when the price is at least a fifth below our estimate, as a cushion for being wrong.
The price and our view · 14 · How could we be wrong?
Flat cars and an AI boom that can turn
Cars are not growingBiggest
Flat for three years.
AI spending can turnWatch
It is someone else's decision.
Light instead of copperShowing now
NVIDIA ships light switches.
Its real AI salesUnknown
The AI figure is not filed.
Tariffs between the US and China hit costs and customers.
The price and our view · 15 · How will we know?
January shows whether cars grow too
Sales of about $5.35B, orders above salesGood sign
Orders below sales, or data-centre sales fallingBad sign
Car sales up 4% or more in fiscal 2027We turn positive
JAN 202720
First-quarter fiscal 2027 results, 20 January. Companies report every three months; each report is a test of the story.
Our view also changes on price alone: we turn positive if the shares fall a fifth below our estimate, and negative if they rise a fifth above it.
The price and our view · 16 · So, what do we think?
A booming AI half and a flat car half
$244our value
Attractive below $195
Neutral
Unattractive above $292
NeutralMedium conviction
TE makes connectors for cars, AI servers and the power grid. Cars, about two fifths of sales, have not grown in three years, while data centres grow fast. The price sits a little below our value. Neutral.
↑ ↓ to move · C for contents
What it does
More about the company
Who it is
Every car, server and power line is full of wires, and every wire ends in a connector: a stamped metal contact in a plastic housing that must carry current or data for years through heat and vibration.
TE makes these, plus sensors and cable, and sells them mostly straight to the manufacturers who design them into their products. It competes in a $99.2B world market in which the hundred largest makers ship 82%. In AI server racks, its copper cables carry data between processors.
How it makes money
TE is paid per part, across tens of thousands of different parts. Half its sales come from cars, trucks and sensors; half from data centres, factories, aircraft, the power grid and medical devices. No single customer matters much. The parts are cheap, but they are designed in years ahead, so a won design pays out for the life of the car or server.
Growth and profit
What growth costs
Growing does not make TE's overheads cheaper: research and selling costs rise about in step with sales. Its margin has widened anyway, because better factories and price rises let each product earn more over what it costs to make. So more sales add profit at a steady rate, not a rising one.
Its record
The numbers
Is it growing?Steady
Sales grew 15% in the year to Sep 2026, faster than the 8.9% in the year before.
Is it profitable?Strong
Operating margin is 20%, up from 14% in the year to Sep 2023.
Does growth pay off?Flat
Running costs take 16¢ of each sales dollar, about where it was (15¢) in the year to Sep 2022.
Does it earn good returns?Decent
Each dollar put into the business earns 18% a year after tax, about where it was (16%) in the year to Sep 2022.
Is it making cash?Strong
It turned 17% of sales into free cash, $3.40B over the year.
Is the balance sheet safe?Light debt
It owes $3.94B more than the cash it holds, about 0.9 years of the cash its operations bring in.
Are shares being diluted?Shrinking
The share count fell 9.5% in 4.0 years, and pay in shares costs 0.8% of sales.
Why it wins
The evidence
Carmakers test a connector through development, production and use in the field under a standard called USCAR-2 before they will build with it. Swapping it later means testing again, so the supplier usually keeps the part until the car is redesigned.
The trace shows up as staying power rather than growth: car sales held at about $7.0B a year while European and American car production fell. TE's own report still admits steady pressure on prices.
Who could beat it
The rivals in full
In AI racks, Amphenol, TE's closest rival, calls itself the market leader and competes for the same designs. NVIDIA already ships switches that carry data as light rather than through copper. TE says copper stays in charge inside the rack, and its own light-based product earns little before 2028.
TE's high-speed copper cable, board connectors and power connectors inside AI server racks; Amphenol calls itself the market leader there and competes for the same designs.
NVIDIA Spectrum-X and Quantum-X Photonics co-packaged opticsNVDA
Copper and pluggable links at the switch; if optics move inside the rack, the copper cable TE sells between processors.
Aptiv Engineered Components GroupAPTV
TE's connectors and interconnects on carmakers' next platforms, sold to the same buyers.
How big it could get
The market and who has signed
The market
The world spent about $99.2B on connectors in 2025, up 14.7%, with data centres the biggest buyer and cars second. TE's sales are at most about 17% of that. Carmakers built 96.4 million vehicles, and TE sells about $73 of car parts for each. Data-centre power capacity is set to grow 31% in 2026.
Who runs it
Leaders, pay and the cash
Who runs it
Terrence Curtin has run TE since 2017, after years as its finance chief, and Heath Mitts has run finance since 2016. Under them the share count has shrunk every year and profit margins have widened since 2023. Insiders own very little, well under 1%, and most of that is options. The yearly bonus pays on sales, operating profit and earnings per share; the long-term award pays on earnings-per-share growth against peers, half of it in options worth something only if the price rises.
What they did with the money
It kept buying back its own shares: $1.35B in nine months, so the share count fell 1.6% since November. It also spends cash on companies: about $2.3B for a grid-connector maker last year and $1.4B agreed for a power-filter maker.
The last report
The latest report
TE Connectivity's sales rose 12% to a record $5.31B and orders reached $6.06B
What it did to the thesis Strong quarter
It met our test: sales, profit and orders beat the guide, the data-centre unit held its level, and cars grew. The car half is still only one quarter of growth, so we stay neutral until fiscal 2027 shows it.
The report
In the fourth quarter of fiscal 2026, ended September 25, sales rose 12% to $5.31B and adjusted earnings per share rose 21% to $3.14, both above the company's own guide. Industrial sales grew 20%, led by the data-centre unit, up 34%. Transportation grew 8%, and car sales grew 7%, or 6% once currency is taken out.
Orders rose 29% to $6.06B, about 14% more than the company shipped. For the first quarter, management expects sales up 15%, and for fiscal 2027 growth like fiscal 2026's 12%.
What's happening now
The background
The price has slipped while the business grew. In April the shares fell 9.1% in a day after the data-centre unit stalled for two quarters; it then jumped in June, and the price barely moved.
TE reported on October 28 and guided the first quarter to sales up 15%. The next check is January, when we see whether orders keep beating sales and whether cars grow 4% or more.
Where we disagree
Both sides in full
The market
A car-parts supplier with an AI side business that may fade as light replaces copper in data centres.
We believe
The AI and grid half keeps growing into 2027 on record orders, but the flat car half caps the whole company.
The argument
The market's worry shows in the numbers: analysts split evenly between buy and hold, and the price fell as sales per share rose.
Where we differ is that power and grid orders are growing too, and light cannot replace a power cable. The evidence against us would be data-centre orders falling, or car sales falling while carmakers build more.
What it's worth
How we value it
Valued on the analysts' revenue path for three years, each later year closing 30% of the gap to 4%, on official operating profit with stock pay a cost and tax paid in full, struck on the fiscal 2026 statements in the 10-K; the thesis's own evidence is the check on that path.
What the price assumes
Sales growth in the first year, then slowing as ours does, at the same profit per sale.
The price assumed
5.9%
Our estimate
12%
Measured at the close of
9 Nov 2026
The three cases
Bear case$136
Growth runs at half the analysts' path, 6%, 4% and 3%, the margin settles at 17.0%, below its 17.6% five-year median, and new capital earns only the 8% rate: the downturn test.
Our value$244
Revenue grows 12%, 8% and 6% over three years, the analysts' path (C51), then fades toward 4%, closing 30% of the gap each year; the margin moves to 21.2%, the best it has filed before amortisation, not the analysts' 22.4% (the roster's cyclical cap).
Bull case$312
Growth runs a quarter above the analysts' path, 15%, 10% and 8%, the margin reaches 23.3%, and new capital earns 15%.
The bet
Seen as a bet
This bet is on one business only: Industrial Solutions, the half that sells into AI data centres, the power grid, factories and defence. The car half is counted separately at the price TE used to trade at, because it has barely grown in three years.
If Industrial keeps its recent pace for two more years, the stock is worth more than today but well short of double, so this is an expensive bet. If that growth stalls, the two halves together are worth somewhat less than today's price.
What could go wrong
Every risk
The risks
Cars are not growing
Car sales, 41% of TE, were flat for three years, and in fiscal 2025 they trailed world production.
AI spending can turn
The data-centre boom is someone else's decision; when data-centre sales paused for two quarters, the shares fell 9.1% in a day.
Light instead of copper
NVIDIA already ships switches that carry data as light; if that moves inside the rack, TE's copper cables lose their place.
A stronger rival in AI
Amphenol calls itself the leader in data-centre connectors and is worth far more per dollar of profit.
China and trade
A fifth of sales are invoiced in Chinese currency, and tariffs between the U.S. and China hit costs and customers.
What we still do not know
How much of TE's AI sales are real. The company gives an AI figure it does not file, and says it will now report only the wider data-centre unit.
Whether TE is winning or losing market share. The research house that ranks connector makers sells the figures, so we could not check.
When, or whether, light moves inside the AI rack. Nobody we could reach publishes a date.
What changes our view
Every route to a new view
Every route to a new view
→ Attractive
The share price falls to about $195, 80% of what we think the business is worth.
→ Unattractive
The share price rises to about $292, 20% above what we think the business is worth.
→ Attractive
Car sales grow 4% or more in fiscal 2027, after taking out currency and purchases. Industrial orders keep growing from a year earlier through fiscal 2027.
→ Unattractive
Data-centre sales fall below the same quarter a year earlier. Orders fall below sales for two quarters in a row.
The next test
It supports the case if
Sales at or above $5.35B, adjusted EPS at or above $3.30, orders above sales with book-to-bill at or above 1.05, digital data networks sales not below the $949M of the September quarter, and automotive organic growth of 4% or more.
It pressures the case if
Sales below $5.25B, book-to-bill below 1.0, digital data networks sales down on the quarter before, or automotive organic growth under 4%.
We would also be wrong if
Operating margin, before one-off charges and amortisation, falls below 20% for a full year.
The share count starts rising, or TE pays for a purchase in shares.
The verdict
Scores and how we weighed it
The scores
Growth
3 of 4
Business quality
3 of 4
Financial position
3 of 4
Execution record
3 of 4
Per share capture
4 of 4
Potential
2 of 4
Risk profile
3 of 4
Valuation
3 of 4
Evidence quality
3 of 4
How we weighed it
ValueMixed
The price is inside the margin of safety, with the bear case well below.
BusinessPositive
Connectors inside cars, AI servers and the grid, a solid and cash-generative business.
GrowthMixed
Data centres and the grid grow fast; cars have been flat for years.
RisksNegative
Cars may stay flat, AI spending can turn, light may replace copper and a stronger rival leads in data centres.
RivalsMixed
Well below its closest rival, which sells more into AI; the gap reflects the mix.
The groupMixed
It sits in none of our ranked groups; car production and AI spending move it.
Latest newsPositive
The latest report beat the guide on sales, profit and orders, and cars grew.
In plain words
Our estimate of value
We forecast how much cash the business can earn over the next ten years and beyond, then ask what that stream is worth today to an investor who wants an 8% yearly return. We do it three times: a gloomy future, an expected one and a sunny one. We call a share Attractive only when its price is at least 20% below the expected estimate, and Unattractive when it is 20% above.