Emerson

IThe business

Emerson sells the equipment and software that run industrial plants. That means control valves, pressure and flow sensors, the control systems that tell them what to do, test equipment, and software that models and tunes a refinery or a power station. Its customers are in power, liquefied natural gas, chemicals, life sciences, semiconductors, aerospace and defense, and a long list of other process and hybrid industries.⁸³ The head office is in St. Louis. At about $95B it is the largest company by market value I found based in Missouri; that is ahead of O’Reilly Automotive of Springfield, at about $75B C (calculated).¹²⁹¹⁰

Emerson’s fiscal year ended on 30 September 2026, and it published results on 4 November. Sales were $19.0B, up 5%, or 4% before currency. The adjusted segment margin was 28.2%, adjusted earnings per share $6.59 and free cash flow $3.5B.² The Form 10-K has not been filed. I use the results release, which is unaudited, and the 10-Q for the quarter to June for detail. Where a number can still move, I say so.

Why it is not a normal industrial business

A plant buys a control system or a valve once and then maintains it for decades. Emerson sells the spares, the repair, the upgrades and the software licenses for that whole life. Of its fourth-quarter sales, 63% was maintenance, repair and operations work, 17% was modernization and expansion of existing plants, and 20% was new plants.³ That is why the business bends in a downturn and does not break. Revenue fell 9% in fiscal 2020, from $18.4B to $16.8B, and passed its 2019 level again in 2022 C (calculated).¹¹¹²¹³

The model has a second leg now. Emerson’s recurring software contracts are worth about $1.7B a year, a figure it tracks as annual contract value, and that grew 10.1% in 2026. The company expects another 10% in 2027.³ Software is the part of this business with the least dependence on what a plant builds this year.

The three groups

Since November 2025 Emerson reports five segments in three groups.¹⁵ Software & Systems is a third of sales and the best business. It holds the control systems (DeltaV and Ovation), AspenTech’s asset-optimization software, and Test & Measurement, which is NI. Intelligent Devices is 55% of sales: sensors and measurement (Rosemount, Micro Motion) and valves and actuators (Fisher, Bettis, Keystone). Safety & Productivity, 11%, is a tool business with brands like RIDGID and Greenlee, sold to tradespeople, and it is the weakest on margin and growth.⁸¹⁵²

Two things in that table matter later. The software group makes the best margin and grew fastest. And NI, which cost $8.2B, shows a 29.0% adjusted margin but only $47M of earnings once its own amortization of $429M is counted.² That is not a fault in the product. It is what the purchase price looks like in the accounts.

A company rebuilt in three years

This is not the company of five years ago. Emerson sold its largest division and its food-waste business, and bought two software and measurement businesses. The table is the whole sequence.

What stands out is the direction. Cash from the sales roughly covered the purchases, and the company that remains is more software, more instruments and fewer compressors. Emerson says the transformation is complete.¹⁵ In August it also said it was pivoting capital allocation to returning cash.¹⁶

Where the revenue comes from

About half of sales are in the Americas, 20% in Europe and 30% in Asia, the Middle East and Africa. In fiscal 2026 the Americas grew 7%, Europe fell 1% and the third region grew 3%.³ China was about 10% of sales in fiscal 2025, or $1.8B.⁸ It remains weak: in the June quarter the Intelligent Devices group sold 8% less there and the software group sold 14% more.¹⁵ In the fourth quarter the United States grew 11% and the Middle East and Africa 19%.³ The company says the conflict in the Middle East is a live risk and assumes it stays within the region.¹⁶³

What happened in 2026, and where almost all the growth came from

This is the part I would not skip. In the company’s own breakdown, the growth verticals (power, liquefied natural gas, life sciences, semiconductors, and aerospace and defense) were 27% of 2026 sales and grew 21%. All the other automation markets, 73% of sales, were flat.³ Doing the sum on the slide, the verticals added about $0.9B and the rest about nothing C (calculated). Total sales rose $1.0B.³² So an investor in Emerson is mostly buying the verticals, and for 2027 the company plans for low double-digit growth there and low single digits elsewhere.³

Of the 4% underlying growth in the year, 3 points came from price.³ In the fourth quarter volume contributed 4 points and price 3, which is better, and orders rose 14%.³ Backlog at year end was $9.6B, or $8.2B before AspenTech, up 11% on the year before.³ That is the evidence behind the 2027 guide of about 5% underlying growth, which on $19.0B is $19.9B of sales.³

The company also gave targets for 2028: $21B of sales, a 30% adjusted segment margin and $8.00 of adjusted earnings per share, with $10B returned to shareholders over 2026 to 2028.³ I use these as the bull case in Section VII and as my base in the model.

It has about 71,000 employees and about 120 plants, 35 of them in the United States.⁸ Research and engineering spending was 8.1% of sales in fiscal 2025.⁸

IIThe moat

The protection comes from where the product sits. A control system runs a plant, and a plant that makes chemicals or power cannot stop to test a new one. A safety valve has to work every time, and a plant manager cares more about the price of a failure than the price of the valve. Emerson’s products are specified into the design of a plant, and the spares, the service contracts and the software follow. Gross margin shows the result: 49.0% in 2023 and 53.4% in 2026, after the sale of the compressor business C (calculated).⁴²

Evidence it is widening

Price is the clearest sign. It added 3 points of sales growth in 2026, and net price was ahead of cost inflation in the fourth quarter.³ Orders accelerated from 7% growth in the June quarter to 14% in the September quarter.¹⁶² Backlog is up 11% before AspenTech.³ The software group’s margin rose to 31.2% from 30.6%, and company-wide adjusted segment margin to 28.2% from 27.6%.² None of that happens in a business whose customers can easily leave.

The cash tells a similar story. Free cash flow was 18.6% of sales in 2026, up from 18.0%, and the company guides above 19% in 2027 and 20% in 2028.³

Evidence it is narrowing

Three signs lean the other way. The first is the split of growth. Price did most of the work in 2026, and when a business grows mostly on price, the question is how long customers accept it. The company itself said volume was held back by about 1.5 points in the first nine months by the timing of software renewals and the conflict in the Middle East, so some of this is timing.¹⁵ But volume was not the engine. The second is concentration. All the growth came from 27% of sales, and the 73% that makes up the rest of the company did not grow.³ The third is geography. Europe and China, together about 30% of sales, were weak in 2026 and are assumed weak again in 2027.³

There is also a competitor that grows faster. ABB took orders up 40% in the September quarter, helped by data centers, to $12.1B, and made a 20.8% operational EBITA margin on $9.7B of sales.¹⁷ ABB is not the same business, because it sells electrical and robotics equipment as well. But it shows where the money is going: to power for data centers, and to those who sell into it.

A word on how I think about the protection around the new businesses. NI sells test equipment that is designed into products. That is a good position, but the segment’s sales fell from about $1.7B when Emerson bought it to $1.46B in 2024 and $1.49B in 2025, and only got back to $1.73B in 2026.⁸² AspenTech’s protection is better, because a plant that models itself in a tool does not switch. I will say it again where I come to management: I like the protection, and I am less sure Emerson paid a fair price for the newest of it.

IIIThe competitive advantage

No two of these companies report margin the same way, so I show each as reported and do not net them. The point of the table is the range of sizes and the direction, and I say below which one is closest to Emerson’s own number.

Where it wins

It wins on breadth in process plants. A refinery or a power station can buy the control system, the safety system, the valves, the instruments and the optimization software from one company, and Emerson is one of few that offers all of it. Its margin on adjusted EBITA after corporate cost, 26.6%, is above Rockwell’s 22% after a similar deduction, and above the 18.7% Siemens Digital Industries reported. The 26.6% includes $125M of pension income, so without it the figure is 26.0% C (calculated).²⁵⁶

It also wins on how much of its revenue repeats. Almost two-thirds of its sales are maintenance work on plants that are already running. Rockwell, whose strength is discrete factories, reports services at about a fifth of sales, and its segment margin in that business was 15.1% in the June quarter, against 34.8% in software and control and 20.0% in devices C (calculated).⁵

Emerson and Rockwell, side by side

Rockwell is the nearest listed name for an investor, and it is the better test of whether Emerson’s price is high or only the sector’s. They are different at the core. Emerson is process, which means chemicals, power and gas. Rockwell is discrete, which means cars, packaging and food. Emerson’s growth in 2026 was 5% and Rockwell’s nine-month growth was 10.5%, with a 9.44 diluted GAAP EPS in nine months that includes a small gain on dissolving a joint venture C (calculated).²⁵ Rockwell has not yet reported its fiscal fourth quarter, so I use the twelve months to June, and Emerson’s year to September. The periods differ by one quarter.

On the numbers, Rockwell costs more for each dollar of earnings and returns less cash on its price. Emerson is the cheaper of the two, and it is still at 29 times owner cash. That tells me the sector is priced for growth, and Emerson is not an outlier within it. I do not take it as a sign Emerson is cheap.

Where it loses

It loses on growth in the businesses that are not the verticals, and on discrete automation, where Rockwell is the stronger name. It loses to ABB on exposure to the fastest part of the market, the electrical equipment that data centers need, because Emerson sells to power producers and does not make the transformers and switchgear.¹⁷ And in software it is smaller than Siemens, which reported software revenue of €1.8B in one quarter, up 15%.⁶ Emerson’s whole software line is about $1.7B of annual contract value.³

None of that makes the business weak. It makes it a specialist with a strong protected core and no obvious route to growing faster than its markets without buying something.

IVManagement

Lal Karsanbhai, 56, has been chief executive since February 2021. He ran the process-automation business from 2018, and before that its measurement unit, so he knows the core. The board chair is James Turley, the retired chairman and CEO of Ernst & Young, an independent director.⁸⁷ The chief operating officer is Ram Krishnan, and Michael Baughman has been chief financial officer since May 2023.⁸

How they are paid, and what that pays them to do

The chief executive’s fiscal 2025 pay was $21.3M, with a salary of $1.54M, a bonus of $3.73M and the rest in stock. That is 402 times the median employee’s $52,916 C (calculated).⁷ Performance shares are the main item, 42% of his pay.⁷

It is the design that matters. The annual bonus pays on adjusted earnings per share (70%) and operating cash flow (30%). In 2025 it paid 141.5% of target.⁷ Performance shares, 55% of the long-term award, pay on adjusted earnings per share (60%) and cumulative free cash flow (40%) over three years, with a modifier for total shareholder return against the S&P capital goods group. The other 45% is time-vested stock.⁷

Look at what those measures leave out. Interest on the debt used for a purchase does flow through both, and so does the interest income given up on cash spent. What neither charges is the amortization of acquired intangibles, which was $1.38 a share in 2026 or about $0.78B after tax, the deal fees, or the capital itself.² So the main things the team is paid on do not count the full price of an acquisition. That is not unusual. But Emerson spent $8.2B on NI in October 2023 and about $7.2B on the rest of AspenTech in March 2025, and neither measure counts the $15.4B itself, only the interest on the debt that helped fund it.⁸ A team paid this way is not pushed to be careful about price, and it has the incentive to buy growth because the growth counts and the cost does not. I do not say that is how they decided. I say the plan would not have stopped them.

The stock modifier helps. Relative return against the S&P capital goods index can raise or lower the payout, and the shares did well: total return was 25% in fiscal 2025 against 17% for the S&P 500, the proxy says.⁷ Targets are set once, at the grant, and not reset, which is better than most.⁷

How much of their own money is in

A fair amount. Karsanbhai held 363,437 shares on 30 September 2025, which includes 75,000 restricted shares, and the proxy puts his holding at about 31 times his salary.⁷ At the close that is about $62M C (calculated). Directors and officers as a group held 1,297,097 shares, under 0.3% of the company.⁷ The policies are sound: officers cannot hedge or pledge, and the CEO must hold six times salary.⁷ A Schedule 13G filed on 29 April shows Vanguard Capital Management with 42,168,982 shares, 7.50% of the company, at 31 March 2026. That is an index manager and not an activist.²⁰

What they did with capital over a full cycle

Over ten years Emerson generated $24.9B of owner cash and returned $21.3B of it, $12.2B as dividends and $9.2B as buybacks C (calculated).¹³² It also made two large purchases and several sales, set out in Section I. The record has two sides.

The sales were well timed. It sold a majority of Climate Technologies in a $14.0B deal in 2023, sold the retained stake and the note for $3.4B in 2024, and sold InSinkErator for $3.0B.⁸ Net debt is lower than it would have been.

The purchases are harder to judge. NI cost $8.2B for a business that earned about $170M before tax and had $1.7B of sales.⁸ In 2026 its segment adjusted EBITA was $503M and its earnings after its own amortization were $47M.² That is a return of 6% on the purchase price before tax, on a favourable view of the adjusted figure C (calculated). The rest of AspenTech cost about $7.2B.⁸ Software & Systems as a whole, which holds both and the older control-systems business, made $1.97B of adjusted EBITA on $6.3B of sales.² I cannot separate what the purchases earn on their own. The one clean case, NI, is a reasonable business bought at a full price.

Buybacks, where the board did decide on price, were made at lower prices than today’s.

The buybacks in 2023 to 2025 cost an average of $103, and the shares are 65% above that C (calculated). They worked. The plan now is to spend about $2.5B in 2027, at a price 1.7 times the earlier average, and to spend $10B on shareholders over 2026 to 2028.³ The company says it is pivoting from buying businesses to returning cash, and the shares are at the highest close of the decade. I cannot say that is wrong. Buying back stock above its value hands some of the value to the sellers, and by my model that is what the price is.

My summary is this. The management has repositioned the portfolio with skill, it has paid shareholders every year, and the CEO owns a lot. The pay plan does not measure what acquisitions cost, and the two largest purchases took the best part of the decade’s cash. I would want the next pay plan to charge for acquired intangibles.

VFinancials

The ten-year view is awkward, because the company changed so much. The table shows every figure as first reported, and the breaks are marked in the notes: Network Power left in 2016 and 2017, Climate Technologies is in the numbers to 2022 and out from 2023, and NI is in from October 2023. The 2024 and 2025 earnings use the recast of 13 October, which changed the pension accounting.²¹ Fiscal 2026 is the unaudited release. I cannot give a clean ten-year organic growth rate, and I will not make one up. For the last two years the company gives 3% and 4% underlying.³

Owner cash against reported earnings

Owner cash is operating cash flow less capital spending less stock pay. I take stock pay out because it is a real cost to a holder, paid in shares instead of cash. Capital spending of $443M is about equal to depreciation, which I take to be about $0.47B, so the business does not need more than that to stand still C (calculated).² In 2026 owner cash was $3.3B. Net earnings were $2.7B, and the adjusted number that management reports was about $3.7B.² The three figures answer different questions. The first charges for the price of acquisitions through amortization. The second does not. The third is the cash, and it sits in between.

Two things in the cash deserve care. Working capital took $338M in 2026, against $9M in 2025, so cash would have been about $0.3B higher without that drag C (calculated).² And the figure includes $82M of tariff refunds received in the June quarter, which will not repeat.¹⁵ The drag is larger than the one-off, so $3.3B is if anything a little low. I take it as the base, and it is 89% of adjusted earnings.

Cash returned

Dividends have been flat in dollars for ten years, at $1.2B, because the share count has fallen as the payout per share has risen. Buybacks have varied: $0.4B to $2.0B. The total over ten years was $21.3B against $24.9B of owner cash C (calculated).¹³² The board raised the quarterly dividend 7%, to $0.595, on 4 November. It was 69 consecutive years of increases at the end of fiscal 2025.²⁷

For 2027 the company plans to return $3.8B, $2.5B in buybacks and $1.3B in dividends, against free cash flow of $3.8B to $3.9B.³ That is all of it. Owner cash after stock pay would be about $3.6B, so the company is planning to return a little more than it earns, and net debt will not fall. In 2026 it fell $1.2B, to $10.4B. That is a fair use of a balance sheet at 1.9 times earnings, and it means any other use of cash, a large purchase for instance, would have to be debt.

The balance sheet

Debt is moderate. Net debt is $10.4B, 1.9 times adjusted EBITDA, and interest takes about 7% of adjusted EBITA C (calculated).² A little over a third of the debt, $4.7B, is short term, mostly commercial paper, backed by $5.5B of committed lines that were undrawn at 30 June.²¹⁵ The point to watch is the asset side. Goodwill and other intangibles are $26.5B of the $41.8B of assets C (calculated).² A business that earns what this one does would not be expected to write them down. If its growth stalled, the question would come up.

The price against owner cash

The share price has risen 245% in ten years, from $49.41 on 4 November 2016 to $170.44, a compound 13.2% a year before dividends C (calculated).¹ From 4 November 2016 to 28 October 2026, the latest day in my S&P data, the index rose about 271% and Emerson’s price 239%, so on price alone Emerson trailed the market C (calculated).²² Owner cash per share has risen 79%, from $3.27 in 2017 to $5.87 in 2026 C (calculated). So most of the price gain came from the multiple, not from earning more.

At $170.44 each dollar of owner cash costs $29. That is a price for what comes next, not for what has been done. The next section asks what it needs.

VIValuation

First, how this goes badly wrong

Before the numbers, I asked what would most likely hurt this business or this investment, and worked back. Three things could, and none needs a surprise.

• The cycle turns after a boom in orders. Orders rose 14% in the last quarter, led by power, gas and chip plants, which are projects.³ Projects finish. A year or two on, a business that is 63% maintenance can still see revenue fall: it did, by 9%, in 2020 C (calculated).¹¹¹² If volume turns down while price has already done its work, the margin that depends on mix goes with it. My bear case has revenue falling 8.5% in 2029, as it did in 2020, and the margin at 25.5%. Even that is worth $75, which is 56% below the price.
• Growth is bought again, with debt. Emerson spent $15.4B on two purchases over 17 months, and one of them, NI, earned $47M in 2026 after its own amortization.⁸² The company now says it is returning cash instead. But the plan returns all of its free cash flow, and the next large purchase would come from debt. If a purchase of $8B turned out to be worth half, that is about $7 a share E (estimate).
• The price falls with no operating miss. At $170.44 the shares cost 29 times 2026 owner cash. Emerson has earned a rising multiple for a decade, 19 times in 2017 to 29 times now C (calculated).¹² If it traded at 20 times next year’s owner cash, about $6.55 a share, it would be $131, 23% below today, with no change in the business E (estimate).³ That is the largest risk on this page, and it is not about Emerson.

The three come to one question. Is this a protected business growing at about 5% a year, with a stable return, or a business that has been bid up as part of a theme, and will go back to what it earns? The first sets my base and the second sets my bear. I cannot tell how much of the price is the theme.

What an owner can take out

I value cash, not the adjusted earnings the company reports. For 2026 the cash after stock pay and after the capital needed to stand still was $3.3B, against $2.7B of net earnings and $3.7B of adjusted earnings.² That figure is $5.87 a share C (calculated). For 2027 the company guides free cash flow of $3.8B to $3.9B, and stock pay of about $0.24B, which leaves about $3.6B, or $6.55 a share on 551M shares C (calculated).³ I do not add back the amortization of acquired intangibles to value. It is not cash, but it is what Emerson paid.

What the price implies

At $170.44 the market values Emerson at $95B on 559.7M diluted shares, the average for the September quarter, and the enterprise at $106B after $10.9B of net debt, pension and asbestos C (calculated).¹²¹⁵ That is 25.9 times adjusted 2026 earnings, 35.6 times GAAP, and 23.5 times the $7.25 guided for 2027.²³ The owner cash yield is 3.5% C (calculated), against the 4.90% ten-year Treasury.²³

What does that pay for? Holding the segment margin flat at 29%, the price needs revenue to grow 10.3% a year for ten years, which is $51B of revenue in 2036, 2.7 times today C (calculated). Even at a 31% margin it needs 9.2%. On my base growth, it needs the margin to reach 46%, against 28% today. The company’s own long-run range is 4% to 7% organic growth.³ The price asks for more than the top of it for ten years.

The result is the reverse of my usual finding. On most pages the price asks for less than the company has done. Here it asks for more than the company’s own plan.

What I can’t judge

Three inputs matter and I can’t pin them down. The first is how much of the 3 points of price will last. Customers accept price in a good year for the plant, and I do not know where they stop. The second is how long the power and gas projects last. Management sees a funnel of $13.0B, and about $8B of it is growth verticals, which is a good sign, but a funnel is not an order.³ The third is the tax value of the amortization. Intangibles amortization is partly deductible, and I do not count that. It may be worth $0.1B to $0.2B a year, or 3% to 5% of cash flow, and the filings do not let me size it E (estimate). So here is a range, and not one number. Read my $113 with the grid.

The discount rate

I use 9.5%. The ten-year Treasury yields 4.90% and Damodaran’s implied equity risk premium was 4.41% on 1 November.²³²⁴ My beta for Emerson on month-end price returns against the S&P 500 is 1.36 over ten years, 1.24 over five and 1.37 over three, which give 10.9%, 10.4% and 10.9% C (calculated).¹²² I take 9.5%, about a point under the lowest of them. My reason is that 63% of sales is maintenance work, and revenue fell 9% in the worst recent year, so the cash flows are steadier than the stock’s record suggests. That is a judgement, and the table shows what it is worth. It is half a point above the 9.0% on my Visa page, because a plant-equipment business is more cyclical than card payments.

Three cases

Bear

Orders cool after the project boom. Revenue grows 3.5% in 2027 and 1% in 2028, then falls 8.5% in 2029, the size of the 2020 fall, as long-cycle projects finish and maintenance is deferred, and recovers to about 3%. The margin dips to 25.5% in the downturn and settles at 27.5%. Revenue reaches $21.9B in 2036.

Base

Emerson does what it has said. Revenue grows 4.8% in 2027 and 5% in 2028, with a 30% margin in 2028, then growth fades to 3% by 2036 and the margin stays near 31%. Revenue reaches $28.2B.

Bull

Power, gas and chip plants keep growing at a double-digit rate for longer, software compounds near 10%, and the margin reaches 34%. Revenue grows 6% to 7% a year through 2031 and reaches $33.2B.

The base is the company’s plan. Revenue of $20.9B in 2028 compares with its $21B target, and the 30% margin is its target. On those, adjusted earnings would be about $8.22 a share, against the $8.00 it gives E (estimate).³ After that I fade growth to 3%. That is slower than the 4% to 7% the company calls through the cycle, because the pricing will be hard to repeat every year. 56% of the base enterprise value is the value after 2036.

The price against the range

$170.44 is 51% above my base of $113, and 18% above my bull of $144. It is 128% above my bear of $75. On the base cash flows, the price returns 7.5% a year, and on the bull 8.6%. My requirement is 9.5%. Neither case reaches it.

Why the market could be right

Two things could make $113 too low. The first is the growth verticals. They were 27% of 2026 sales and grew 21%. If they kept that pace for ten years and the other 73% grew 3%, total sales would grow 10.3% a year, which is what the price needs at a 29% margin C (calculated).³ I do not expect it, but it is arithmetic, and orders up 14% and a $13.0B project funnel point that way for the next two years.³ The second is the discount rate. The price returns 7.5% a year on my base cash flows. That is a fair return if the shares move about 0.6 times as much as the market, using the 4.90% Treasury and a 4.41% equity premium C (calculated).²³²⁴ I measured 1.2 to 1.4. But a buyer who sees 63% of sales as maintenance work can fairly expect a calmer ten years than the last ten, and I cannot show that buyer is wrong. Leaving out the tax saving on amortization, which I cannot size, makes my cash flows a little low as well.² None of this makes $170.44 wrong. It means the price needs the verticals to keep their 2026 pace for a decade, or investors to accept a return near 7.5%, and I cannot rule out either.

The margin of safety

I set the margin-of-safety price at $100, from my bear case: $75 is at least three-quarters of $100. At that price my base cash flows return 10.2%, above the 9.5% I require, and I lose about a quarter if the bear case comes. My base value of $113 is the price at which I would earn the 9.5% exactly. The stock last closed below $100 on 22 April 2025, at $99.79.¹ At $170.44 the price is a bet that the business does better than the company’s own plan. I cannot say that is wrong, only that the price already pays for it.

Quality and price

These are two questions, and they have different answers. The business is of high quality: a gross margin of 53%, free cash flow of 18.6% of sales, a product that sits in plants for decades, and a company that had raised its dividend for 69 years running at the end of fiscal 2025.²⁷ The price is another matter. Emerson looks like a very good business at a full price. A good business at a full price tends to earn what the business earns, less the premium paid, and here the premium is 51%.

The next best use of the dollar

A Treasury note pays 4.90% for ten years, with no risk to the principal if held.²³ The yearly return on my base cash flows is 7.5%, which is 2.6 points more, for carrying the risk of a company. That is thin. The nearest listed alternative is Rockwell, and I compared the two in Section III: its owner cash yield is lower and its price over GAAP earnings higher, so the same dollar does no better there.⁵¹⁸ On my base cash flows Emerson pays only a few points over the safe rate, and I would want more for the risk of a company.

VIIThe case for the price

What most people miss

Three facts cut in Emerson’s favour and get little attention. First, of fourth-quarter sales, 63% was maintenance work on plants already running.³ Second, in 2026 the growth verticals, power, gas, life sciences, chips, and aerospace and defense, were 27% of sales and grew 21%, and the other 73% was flat, so the company is a faster-growing business than its total shows.³ Third, the order book is moving: backlog before AspenTech was up 11% to $8.2B, fourth-quarter orders were up 14%, and project wins of about $575M were about 75% from the verticals.³

The bull case

In plain words: Emerson sells the control layer for the plants that electrification, energy security and chip making are building, and the margin rises on each added dollar of sales. The clearest statement comes from management. On 4 November Lal Karsanbhai, the chief executive, said he has confidence in 2027 given backlog and demand in the growth verticals.² The company published numbers to go with it: $21B of sales in 2028, a 30% adjusted segment margin, $8.00 of adjusted earnings a share, a 20% free cash flow margin, and $10B returned to shareholders over 2026 to 2028.³ Adjusted earnings of $8.00 in 2028 is about 10% a year from $6.59.³

I looked for a named investor or analyst who has published a view on Emerson in the past year and could not verify one in a primary document, so I leave the case to the company. The one large holder I could verify on a filing, Vanguard Capital Management, with 7.50%, holds the shares for its funds and clients, and I found no view from it on the company.²⁰

Against reality

My base is this plan. It has sales of $20.9B in 2028 against the $21B target, a 30% margin, and about $8.22 of adjusted earnings a share against $8.00 E (estimate).³ So if Emerson does what it says, the value is about $113, and not $170.44. The price asks for more than the plan. Over ten years the price needs 9.2% growth a year at a 31% margin, or a margin of 46% on my growth C (calculated).

The evidence for the plan is better than for most. Backlog is up and orders accelerated. The record against guidance is good but not perfect: underlying growth ended at 4%, the first guide, after August’s cut to 3.5%, and adjusted earnings of $6.59 beat the first range of $6.35 to $6.55. Free cash flow of $3.5B was the bottom of its first range and below the $3.6B of August.³ The weak point is the mix. Price gave 3 of the 4 points in 2026, and only the verticals grew.³ A plan that needs volume to carry the next two years has not yet had that test.

It helps to multiply it out. My base needs four things. Volume recovers so that growth stays near 5% through 2028: 70%. The margin reaches 30% and holds: 75%. No large debt-funded purchase: 85%. No downturn deeper than 2020 before 2030: 70%. Those are my guesses, and they multiply to 31% E (estimate). They overlap, so perhaps 40% in practice. The price needs far more than that, for a decade, and I put it under 10% E (estimate).

What a 10x would take. Ten times $170.44 is $1,704, about $954B of market value on today’s shares. At 25 times earnings that needs about $38B of net income, ten times the $3.7B of adjusted earnings in 2026. At a 20% net margin, just above the adjusted 19.5% of 2026, that is about $191B of revenue, ten times 2026. By 2036 it needs 26% growth a year for ten years, from a company growing 5%. Reported revenue grew 24% in nine years, from $15.3B to $19.0B, and part of that was acquisitions C (calculated).¹⁴² Buybacks of 2.5% a year would cut the revenue needed by about a fifth, to about $149B, which is still ten years of growth well above the plan. The one case I can show of a tenfold rise from this size in my price files is NVIDIA, from about $260B on 29 July 2020 to about $2.6T on 24 May 2024, on about 24.6B split-adjusted shares and a surge in chip demand C (calculated).²⁵²⁶ I know of no industrial company that has done it. For Emerson, tenfold from here is close to impossible.

Where I come down

The orders moved me: 14% in a quarter, with the project wins mostly in the verticals, is the strongest evidence yet for the plan. The price did not, and neither did the pay plan, which does not charge for the capital used in acquisitions. Volume growth of 3 points or more for four quarters, with the margin held near 29%, would raise my values most, and a lower price would raise my return.

VIIIRisks, and what would change my view

The three ways this goes badly wrong are at the start of Valuation. These are the rest.

• Weak regions. Europe, about 20% of sales, fell 1% in 2026, and China, about 10% of sales in 2025, is weak again.³⁸ The company plans both flat to low growth in 2027.³ A third region, the Middle East, is the fastest grower and the least certain: the guide assumes the conflict stays within the region.³
• Price against volume. Price gave 3 points of 2026 growth.³ If customers push back, or inflation falls, price falls and volume must replace it.
• Tariffs. The Supreme Court ruled in February that the emergency-powers tariffs were not allowed, and Emerson received $82M of refunds in the June quarter. It does not know whether more will come, and it books none in advance.¹⁵ Other tariffs remain.
• Software-defined and open systems. The 10-K lists artificial intelligence and shifts in technology among the risks that can change what customers choose.⁸ A move to open systems lowers the cost of replacing a control system, which is what the protection rests on.
• Accounting choices. In October the company changed its pension accounting, and adjusted earnings now exclude tariff refunds and the pension mark.²¹² The change was small. I mention it because adjusted earnings carry the pay plan and the guidance, and the list of what is adjusted away has grown.
• Short-term debt. $4.7B of debt is short term, mostly commercial paper. It is backed by $5.5B of lines and the rate is low, but the cost resets quickly if markets tighten.²¹⁵
• Asbestos and pensions. Emerson carries an asbestos liability of $119M and a pension and retiree health liability of $453M.¹⁵ Both are small beside the cash flow, and I deduct them as debt.
• Key people. Karsanbhai has led since 2021 and the plan is tied to him. The 10-K lists the loss of key employees as a risk.⁸