Cencora

IThe business

Cencora is the company behind a pharmacy’s back room. It buys drugs from manufacturers, stores them, and delivers them, to pharmacies, hospitals, clinics, oncologists, veterinarians and others. The name changed from AmerisourceBergen in 2023.¹⁵ Revenue in fiscal 2025, which ended 30 September, was $321.3B and the company had more than 51,000 employees.² Pharmacists and doctors rarely think about it, and that is the point. When the drug is not on the shelf, they notice.

The idea is old and simple. A manufacturer would rather ship to a few warehouses than to tens of thousands of pharmacies. A pharmacy would rather order everything from one supplier once a day. The distributor sits between them, carries the inventory and the credit risk, and keeps a thin slice. Cencora’s slice was 3.57% of revenue as gross profit in fiscal 2025, and 1.31% after the cost of running the business, on an adjusted basis C (calculated).²³ It works because the volume is enormous and the cost per package is low.

How it is organised now

In fiscal 2026 the company changed how it reports. There are two segments and a bucket for businesses it is looking to sell or run differently.⁶ U.S. Healthcare Solutions is the drug distribution business, plus its physician specialty networks. International Healthcare Solutions is Alliance Healthcare in Europe, the global specialty courier World Courier, Innomar in Canada and parts of the consulting business PharmaLex. Other holds MWI Animal Health, the Brazilian distributor Profarma, and the rest of PharmaLex. A fourth piece, the U.S. consulting business, was divested in April 2026.⁶

U.S. distribution

The core is a national network of distribution centers that supplies brand-name, specialty and generic drugs, over-the-counter products and home health supplies to retail pharmacies, mail-order pharmacies, hospitals and health systems, long-term care and physician offices.² Generics are about 90% of US prescription volume but a small share of revenue, since they cost so little.² The company calls the rise of generics and biosimilars favourable because they have given it a larger gross margin than brands.² It sources generics globally, runs a private-label generics program out of Ireland, and gets help from the Walgreens sourcing arrangement described below.²

The growth in dollars comes from two places. One is the unit volume of drugs, which keeps rising as the population ages. The other is the cost of new drugs. IQVIA expects US pharmaceutical sales to grow about 8.4% a year from 2024 to 2029.² The distributor sees that in revenue but not in profit. Its fee is a small percentage of a bigger number, so a $90,000 cancer drug can earn it less per dollar than a $9 generic.

The weight-loss drugs

GLP-1 drugs for diabetes and weight loss are the clearest example. Sales of the class rose $7.7B, or 26.9%, in fiscal 2025, which puts the year’s total at about $36B, or 11% of revenue C (calculated).² In the first nine months of fiscal 2026 they rose another $5.2B, or 19.7%, to about $32B C (calculated).⁶ The company says these products have lower gross profit margins and that they held back the segment’s margin.²⁶ So the biggest source of revenue growth in the last two years is among the least profitable. I keep that in mind every time I see a revenue growth figure on this page.

Physician networks: the new centre of the strategy

The part that has changed most is specialty. Cencora has bought management service organizations, or MSOs, which run the business side of independent physician practices: billing, purchasing, technology, clinical trials. The doctors keep control of care. Cencora gets a management fee, a closer link to the practice’s drug purchasing, and a seat in the fastest-growing part of pharmacy, which is drugs given in a doctor’s office rather than picked up at a counter. I read the company’s statements as saying the goal is to deepen relationships with community providers and build on its leadership in specialty drugs.²

It is writing very large cheques. In January 2025 it bought 85% of Retina Consultants of America for $4.0B in cash, plus $694M of contingent value for the physicians who kept 15%, a $546M receivable settlement and $393M of earn-out, $5.7B in all.⁶ In February 2026 it bought the majority of OneOncology, the oncology network, for total consideration of $7.4B, of which $4.65B was cash and $1.93B was the fair value of the stake it already held.⁶ In March 2026 it agreed to buy EyeSouth Partners’ retina business for $1.1B.⁹ On 21 October 2026 it agreed to buy American Oncology Network, which has more than 350 providers in 21 states, at an enterprise value of $2.4B, or about $1.7B of cash if rollover holders keep 25%.⁵ Added up, that is about $16.6B of purchase price or enterprise value committed in under two years, against a company whose market value is $64B C (calculated).

OneOncology and AON are both in oncology. The deals raise gross margin in the segment. U.S. Healthcare Solutions’ gross margin rose from 2.44% to 2.96% of revenue between the first nine months of fiscal 2025 and 2026, and the company says the main reason is the two networks, offset by the lower-margin weight-loss drugs.⁶ That is real. It is also bought. I look at the price paid, and what the physicians who rolled over keep, in Management and Valuation.

International: Alliance Healthcare and the couriers

Alliance Healthcare supplies pharmacies, doctors, health centres and hospitals in Europe. Cencora bought the majority of it from Walgreens Boots Alliance in June 2021 for about $6.6B in cash plus two million Cencora shares.²¹ Its revenue was $24.4B in fiscal 2025, up 5.8%.² International Healthcare Solutions as a whole earned $588M of adjusted operating income on $28.3B of revenue in fiscal 2025, a margin of 2.07%, and the figure fell 10% from $657M the year before.³ World Courier moves temperature-controlled and clinical-trial shipments for drug makers, and the 10-K calls it a leading global specialty logistics provider.² One part of this segment is not like the rest. A substantial majority of Alliance’s bank borrowings at 30 June 2026 were in Türkiye, and the company has booked $38M of expense in nine months for the highly inflationary Turkish lira.⁶ In the third quarter, the segment’s profit rose 21%, and the company attributes it mainly to the European distribution business, driven by manufacturer price increases in a developing market, plus specialty logistics.⁶ That is a lift I would not extrapolate.

Animal health, and why it is for sale

MWI Animal Health sells drugs, vaccines, parasiticides, diagnostics and feed ingredients to vets and producers; it had $5.7B of revenue in fiscal 2025.² On 17 February 2026 Cencora agreed to merge MWI into Covetrus, a veterinary technology and services company. The deal values MWI at $3.5B. Cencora gets $1.25B in cash, $800M of perpetual preferred units, and $1.45B of common units in the combined company, a non-controlling 34.3%.⁷ The Federal Trade Commission asked for more information on 20 May 2026, so closing is not certain.⁶ The carrying value of the MWI group was $3.8B of assets and $0.9B of liabilities at 30 June 2026.⁶ I can’t tell what a 34.3% stake in a private, leveraged veterinary distributor is worth. The preferred units pay in kind, not in cash. I keep MWI in my numbers at roughly a fair price and say so in Valuation.

Walgreens

Walgreens is more than a customer. It is a former owner, a sourcing partner and the buyer of a quarter of everything Cencora sells. The relationship started in March 2013 with a 10-year distribution agreement, a generics sourcing arrangement and a stake in Cencora that was still about 10% as late as August 2024.²² In June 2021 Cencora bought most of Alliance Healthcare from Walgreens and both agreements were extended by three years to 2029; a new agreement to supply Boots in the United Kingdom runs to 2031.²¹ Walgreens Boots Alliance was taken private by Sycamore Partners on 28 August 2025.² Its holding in Cencora, 19,980,000 shares or about 10% in August 2024, was 12,570,000 shares or 6.5% in February 2025, after a block sale and a share repurchase by Cencora, and 9,615,000 shares or 4.96% by June 2025 as forward sale contracts settled; its board nominee left in May 2025.²²²³²⁴ Today Walgreens is a customer and a counterparty, not an owner.

Three facts about the exposure. Walgreens and Boots together were 25% of fiscal 2025 revenue, about $80B, and 38% of accounts receivable, about $9.6B C (calculated).² The Walgreens US distribution agreement and the sourcing arrangement have stated terms that run to 2029, and the Boots agreement to 2031.² And the 10-K warns that the new owners may expand the plan, announced in October 2024, to close about 1,200 US stores over three years, or change the relationship.² I don’t know what Sycamore intends. The filings don’t say. The company also does not say what the margin on Walgreens is, so I can only size it two ways. At the US segment’s average margin of 1.15%, a 20% fall in Walgreens volume would cost about $0.18B of operating income, about 4% of the total E (estimate). At a contribution margin above the average, because fixed costs stay, it would cost more. I use the larger fall in my bear case.

The second large customer is Evernorth Health Services, at 13% of revenue and 5% of receivables.² In the third quarter of fiscal 2026 the company said sales to its large mail-order customer were lower because of brand conversions, and an oncology customer was lost in June 2025.⁶² Sales to the top ten customers were 66% of revenue.² A distributor does not set its own price against customers this large. I see the margin staying thin partly for that reason.

IIThe moat

A durable advantage is a cost or a habit a rival with money and talent can’t copy in a reasonable time. Cencora has some of both, but they are shared. There are three national distributors in the United States, and the company’s own 10-K names McKesson and Cardinal Health as its largest competitors.² An advantage that three firms share protects the group from outsiders. It does not protect any one of them from the others or from its biggest customers.

What the history says

The best test of an advantage like this is whether the margin held. Adjusted operating income as a share of revenue was 1.38% in fiscal 2016, 1.14% in 2019, 1.33% in 2022 and 1.31% in 2025 C (calculated).^3031324 It has stayed between 1.14% and 1.38% for ten years, while revenue more than doubled. That is a business whose economics hold, and one that does not widen. The profit growth came from selling more, from adding services, and from buying other companies, not from a better price on each dollar.

The same table shows the weakness. Gross margin rose from 2.91% in 2016 to 3.57% in 2025, but part of the 2025 figure was one-off: antitrust settlement gains of $236M and a LIFO credit of $77M, both of which the company excludes from adjusted numbers.² Take them out and the adjusted gross margin was 3.49% C (calculated).⁴ A business that keeps 1.3 cents of each dollar has little room for a pricing mistake.

Two gaps in the protection

Two parts of the protection are weaker than they look. The first is the customer side. Walgreens, Evernorth and the other large chains are far larger than any single pharmacy, and the 10-K itself says key contracts may be renewed on less favourable terms.² The second is the manufacturer side. The 10-K says that an executive order on most-favoured-nation pricing seeks to cut what manufacturers charge in the US, that the government is encouraging direct-to-consumer programs, and that some drug makers already ship direct to patients in ways that may skip wholesale distribution.² That is not today’s revenue. It is the one policy path that goes around the protection rather than over it.

What I make of it: the protection is real, the same width as the other two distributors, and best described as keeping the margin from falling rather than letting it rise. It deserves a low multiple of earnings compared with the businesses on this site that have pricing power, and I carry that into my values.

IIIThe competitive advantage

The industry’s protection covers the group. The advantage is what Cencora has that the other two don’t, and what they have that it doesn’t. The direct comparison is McKesson. It reports on a March year, and it is bigger than Cencora, more profitable per dollar of revenue and priced higher. I took every McKesson figure here from its own filings and releases.^11331229

Where McKesson is better placed

It earns more on each dollar: 1.63% adjusted against 1.38% C (calculated). Most of the gap comes from one segment Cencora has no real match for. McKesson’s Prescription Technology Solutions earned $1.13B of adjusted operating profit in fiscal 2026, at a margin of 19.45%, which is 16% of McKesson’s adjusted segment profit from 1.4% of its revenue C (calculated).²⁹ It sells access and support services to drug makers, the very business Cencora says it wants more of. McKesson also has its older oncology and retina networks, The US Oncology Network and PRISM Vision, and its Oncology and Multispecialty segment earned $1.44B on $48.4B of revenue.²⁹ It has retired a larger share of its stock, 7.5% against 4.6% over two years, and its net debt is lower. And it has been simpler: it finished selling its European business in January 2026 and is separating Medical-Surgical, after selling about 13% of that business to Apollo for $1.25B.¹²²⁹

Where Cencora is better placed

It is cheaper: 19.7 times last-12-month adjusted EPS against 24.5 C (calculated), and 17.4 times adjusted operating income against 19.3 on a like-for-like basis C (calculated).¹³⁴ Its U.S. segment, with its specialty networks, earned 1.15% in fiscal 2025 against 1.03% for McKesson’s North American Pharmaceutical segment in fiscal 2026, though the two segments don’t contain the same things.³²⁹ Cencora has Europe, which lifted international operating income 21% in the June quarter, and World Courier, a specialty logistics business for drug makers. And its animal health sale puts about $3.5B of value in front of a company worth $64B.⁷

What they share

Both depend on one customer for about a quarter of revenue: Walgreens and Boots at Cencora, CVS at McKesson.²¹¹ Both pay opioid settlements over many years, $4.2B at Cencora and $5.66B at McKesson.⁶³³ Both are buying physician practices in oncology and retina, and Cencora says its weight-loss drug sales carry low margins.⁶²⁹ So I treat the pair as one idea with one set of policy risks, not as two independent bets. If drug pricing policy squeezes distribution margins, it hits both. If it doesn’t, the question is which one the market has already paid for more, and on every ratio in the table that is McKesson.

What I read from it

Cencora’s adjusted margin is the lower of the two, its multiple is lower than McKesson’s, and its buyback has been much smaller. None of that tells me that it is mispriced. A lower multiple for a lower-margin business with a larger single-customer risk and a more cluttered balance sheet is what I would expect. What I notice is the opposite problem: even the lower multiple of the two is, on my numbers, above my base value, which is the point of the next sections.

IVManagement

I follow the incentives first, then the money, then what management did with capital over a full cycle.

Who runs it

Bob Mauch became president and CEO on 1 October 2024. He founded the consulting firm Xcenda, which Cencora bought in 2007, ran the distribution business, and was chief operating officer from October 2022.¹³ He followed Steve Collis, CEO for 13 years, who stayed on as executive chairman for a year and retired from the board on 30 September 2025. Mark Durcan, a director since 2015, became independent chair on 1 October 2025.¹³ The chief financial officer, Jim Cleary, who joined with the MWI purchase and held the job from November 2018, retired on 30 June 2026. His successor, Eva Boratto, was most recently CFO of Bath & Body Works and before that CFO of CVS Health, McKesson’s largest customer.³⁵³⁶¹¹ She received a $2.0M sign-on bonus that she must repay on a sliding scale if she leaves within a year.³⁶ A new head of human resources also arrived in July 2026.¹⁰ In twenty months the company has changed its CEO, chair and CFO, in an orderly way according to the filings. I read that as a risk worth watching, not a problem.

How they are paid

The annual bonus pays 40% on adjusted operating income, 25% on adjusted EPS, 25% on adjusted free cash flow and 10% on corporate responsibility goals.¹³ The goals were set in November 2024. They called for adjusted operating income growth of about 4.7%, adjusted EPS growth of about 7.6%, and adjusted free cash flow of $2.0B, which was 34.6% below the year before.¹³ Results were adjusted operating income growth of 15.8%, adjusted EPS growth of 16.3% and adjusted free cash flow of $2.97B C (calculated).⁴¹³ The payout was 181.3%. The committee cut it from 185.3% because of the underperformance of one international business.¹³ The Retina Consultants of America deal closed in January 2025, after the goals were set, and the proxy doesn’t say whether goals were adjusted for it. I don’t know, and I note that it paid the maximum on two of the three goals.

Sixty percent of equity awards are performance shares and forty percent are restricted units. The performance shares pay 75% on adjusted EPS growth over three years and 25% on average adjusted return on invested capital, and a relative-return modifier of plus or minus 15% against the S&P 500 Health Care Providers and Services index.¹³ Half of earned shares must be held for two years by the CEO. If the three-year total return is negative, the payout is capped at target.¹³ Those are good design points. The weakness is what adjusted means. Adjusted EPS excludes amortization of acquired intangibles, deal and integration costs including the equity-unit charges for the physicians who kept a share of RCA and OneOncology, restructuring, and impairments.¹⁰ So management is paid on a number that does not feel the cost of an acquisition that disappoints, and the return-on-capital test counts goodwill but only after-tax adjusted income. It pays them to buy growth.

Their own money

Mauch held 114,546 shares, units and options vesting within 60 days at 30 November 2025, worth about $39M at $337.84, or 29 times his salary C (calculated).¹³ His guideline is six times salary. All 15 directors and executive officers held 743,823 shares, under 1% of the company.¹³ The largest holders were Vanguard at 11.6% and BlackRock at 8.4%.¹³ Among the well-known investors I checked at 30 June 2026, I found no position at Berkshire Hathaway, Altimeter, Atreides, ARK, Akre, Fundsmith, Baupost, Third Point or the Gates Foundation Trust; Wellington Management held 4,559,761 shares.³⁷ Insiders have skin in the game, but not a great deal of it.

What they did with capital

Over ten fiscal years Cencora generated $27.3B of operating cash flow and spent $4.5B on capital, $14.9B on acquisitions, $8.0B on buybacks and $3.6B on dividends C (calculated).²⁰ It also paid or accrued the opioid claims. Here is what the larger purchases look like years later.

The record is mixed. MWI came back at about 1.3 times its price after eleven years, before the profit it made along the way. PharMEDium was a write-off. PharmaLex is mostly written off after two years. Alliance did what it was bought to do, a European platform with Walgreens as anchor, but the returns on $6.9B are not visible in a segment that earned $588M. The physician networks are too new to judge. In all, about $20B has gone into acquisitions over ten years and nine months, counting 2026 so far C (calculated).²⁰⁶ I am more comfortable when a management disposes of what it overpays for, and this one does. On the write-offs in its own filings, it has also overpaid at least twice.

What I can’t judge

I can’t see inside the MSOs. The company does not give their profit, their margins or the terms of the physicians’ equity, other than the liability figures. I can’t tell whether Mauch will treat the physician networks as a tool for selling drugs or as a business to be earned on its own. And I can’t judge the leadership in the finance chair, since Boratto started in June. What I can judge is that the pay design rewards size and adjusted growth, and that the history of buying has been uneven. That is why my base case assumes a return on new capital of 12%, below the 16% in my bull case, rather than the higher return on the old distribution business.

VFinancials

Ten fiscal years, 2016 to 2025, from the 10-Ks and the Q4 releases, then the first nine months of fiscal 2026 from the June 10-Q. The company’s 2026 fiscal year ended on 30 September 2026 and its results come on 4 November, after the cut-off for this page.⁵

Revenue and profit

Revenue grew 9.1% a year over nine years, from $147B to $321B. Adjusted operating income grew 8.5% a year, to $4.22B, and adjusted EPS grew 12.3% a year, to $16.00 C (calculated). The three lines tell the story: revenue grew faster than profit, profit grew more slowly than EPS, and the gap is buybacks and a lower tax rate. Diluted shares fell from 226M to 195M, and the US tax cut of 2018 lifted adjusted EPS that year by cutting the rate.³⁹²

The adjusted profit line had a flat stretch and a steep one. From 2016 to 2019 adjusted operating income was about $2.0B each year. From 2019 to 2025 it doubled to $4.22B, and the growth came from the Alliance Healthcare purchase in 2021 (International), from the specialty business, from the Retina Consultants purchase in 2025, and from drug price inflation.³¹² Over the last five years adjusted operating income grew 13.9% a year C (calculated). Management’s long-term target of 7% to 10% is lower than that, but the last five years included $11.3B of cash acquisitions C (calculated).²⁰

Adjusted against reported

The two measures differ by a lot. Fiscal 2025 diluted EPS was $7.96 as reported and $16.00 adjusted. The company excludes real things for good reason and some it shouldn’t. I read each line below.

Three of those lines recur, and I would not call them one-off: restructuring ($230M, $234M and $229M in the last three years) and the legal fees on opioid claims, and the charges for physician-held units at RCA and OneOncology, which will keep growing as the networks do.²⁶ I treat $200M a year before tax as an ongoing cost in my owner-cash numbers E (estimate).

Owner cash

A reader can’t spend adjusted EPS. What an owner can take out is cash after stock pay and after the capital needed just to stand still. Here is that for fiscal 2025.

So the number I use for what the business yields today is $2.79B, or about $14.3 a share, and the yield on the price is 4.3% E (estimate). That is less than the 4.91% ten-year Treasury yield, with the growth still to come.¹⁴ It also means I discount the adjusted number by about eleven percent before I start.

Suppliers’ money

The balance sheet has a strange shape. Receivables were $25.2B, inventory $20.5B and payables $54.7B at the end of fiscal 2025.² Payables are bigger than receivables plus inventory by $9.0B C (calculated). Suppliers, in effect, finance the stock and the credit given to customers, and the gap grew from $4.0B at the end of 2016 to $9.0B at the end of 2025. That is the float of a distributor and a source of cash while volume grows. In 2025 the working capital accounts supplied $0.44B of cash from operations C (calculated).² On quarterly averages the company counts 27.9 days of receivables, 27.0 of inventory and 59.6 days of payables.² A change in a big customer’s payment terms or a big supplier’s would move cash by billions in one quarter. The 10-K says so.² I count part of the float in my valuation, $4.0B growing with operating income, and put a $2.5B cash hit in my bear case E (estimate).

Where the cash went

Over ten years the company bought back $8.0B of shares and paid $3.6B in dividends, against $14.9B on acquisitions C (calculated).²⁰ In fiscal 2016 it used the $2.36B Walgreens paid to exercise warrants to buy back stock.¹⁹ In fiscal 2026 it paused buybacks in December 2025 to finance OneOncology, then resumed in the June quarter, buying $1.0B of shares.⁸⁶ The dividend is $0.60 a quarter, raised 9% last November, a 0.71% yield.⁴

Debt and other claims

The balance sheet changed in a year. Debt was $7.7B at the end of fiscal 2025 and net debt $3.3B C (calculated). By June 2026 net debt was $8.9B, after the OneOncology purchase was funded by $3.0B of new notes at 3.95% to 5.65%, a $1.5B term loan and cash, and net interest expense is guided to about $490M for the year, against $292M in 2025.²⁴³¹⁰ Management said it would prioritise paying down debt, and it repaid $1.1B of loans.⁸⁶ Then on 21 October it agreed a $2.4B purchase. The company still says it is committed to its investment-grade ratings.⁸

The opioid settlement is the longest claim. The accrual is $4.2B and the company says it expects to pay it over 13 years; payments were $404M in fiscal 2025 and $506M in 2024.⁶² The company recorded a $102M reduction of the liability in the June quarter after the dismissal of opioid litigation, and the City of Baltimore dropped its claims in May 2026.⁶ But the Fourth Circuit sent the West Virginia case back for a new ruling, the Department of Justice’s Controlled Substances Act suit in Pennsylvania has fact discovery to November 2026 and expert discovery to June 2027, and the company says it cannot estimate the range of loss for matters it has not accrued.⁶ I put $1.5B more in the bear case and leave the base alone.

The current year

The first nine months of fiscal 2026 brought revenue of $249.0B, up 4.8%, adjusted operating income of $3.56B, up 11.4%, and adjusted EPS of $13.31 against $12.15, up 9.5% C (calculated).¹⁰⁴⁴⁴⁵ The company raised its adjusted EPS guidance in May and August, and on 21 October said it expects at least $17.90, near the top of the $17.75 to $17.95 range given in August.⁴⁵¹⁰⁵ Revenue guidance is 4% to 6%, adjusted operating income growth 13% to 14%, adjusted free cash flow about $3.0B and capital spending about $900M.¹⁰ Fourth-quarter EPS has to be at least $4.59 for that to hold, 20% more than the $3.84 of a year ago C (calculated).⁴ It is the first thing I will look at on 4 November.

VIValuation

Start with what would go wrong

Before any numbers I ask what would most likely make this a bad investment from here, and work back from that. The business is not about to fail. The three ways it goes badly wrong, in the order I think most likely, are these.

1. The growth that is being bought earns too little. About $16.6B is going into physician networks. If the new capital earns 8% a year instead of my 12%, my base value falls from $243 to $166; at 6% it falls to $90, almost my bear case E (estimate). Management and its pay plan are built to grow adjusted EPS. Whether this growth pays is a separate question and the filings don’t answer it.¹³⁶
2. Walgreens shrinks or reprices. Walgreens and Boots are 25% of revenue and 38% of receivables, and the agreements run to 2029 and 2031.² Take six points off growth in fiscal 2029 and four in 2030, on top of my base, and the base value falls to $226 E (estimate). A contract lost outright is worse than that.
3. Policy and direct-to-patient sales squeeze the 1.3 cents. One tenth of a percentage point off the margin on revenue is about $0.33B, or 7% of operating income, and takes my base to $221 E (estimate).² Drug pricing orders, direct-to-consumer programs and the pace of brand price inflation all point that way. I can’t put honest odds on the path, so I don’t.

I can show what each does to the numbers, and I do below. I also keep a fourth possibility in mind that is not a business risk at all: that the multiple falls while the business does fine, as it has this year, from 21.1 times to 18.9, and fell to 12.0 times in 2019.

What the price implies

In October 2025 the shares closed at $337.81 and adjusted EPS for the year just ended was $16.00, which made 21.1 times.¹⁴⁶⁴ Today they close at $337.84 and the company expects at least $17.90, a 12% rise, which makes 18.9 times C (calculated).¹⁵ Nothing broke. Earnings went up 12% and the price went nowhere, so the multiple fell about two points. The longer record is the other way: it rose from 12.5 to 21.1 in nine years, and it was 12.0 in 2019 C (calculated). Since 2019 the price rose about four times: adjusted EPS rose 2.3 times and the multiple 1.8 times C (calculated). Margins on each dollar did not rise.

Reverse DCF: what the price needs

Held at a 12% return on new capital, the price needs adjusted operating income to grow about 11% a year for ten years, to $13.8B in fiscal 2036, against my base of 5.8% a year and $8.5B C (calculated). The past is a mixed guide. Adjusted operating income grew 12.8% a year from fiscal 2019 to 2025, but that included $11B of purchases, and the same measure was flat for the three years before.²⁰ If new capital earns 25%, the growth the price needs falls to 6.6% a year C (calculated). The target management gives is 7% to 10%.⁸ At 10% a year for ten years with a 12% return, my model gives $313; with 25% it gives $450 E (estimate). So the price is reasonable if management hits the top of its own range and the money it spends earns twice what I assume. It is not reasonable on the middle of the range with the returns the acquisitions have shown so far.

There are two other ways to read the same price. One is that it equals my base cash flows at a discount rate of about 7.3% C (calculated). That is the return the price offers if my base is right, 2.4 points above the Treasury. The other is to ask what a buyer in 2036 pays. The table below holds my base earnings and changes the multiple.

If the market still pays 17.3 times in 2036, my base earnings give 10.4% a year, above my required rate. If it pays what my base value implies, 11.2 times, the return is 6.0% E (estimate). The 7.3% above is the rate at which my base cash flows, with the value after 2036 set at that same lower rate, equal the price; it assumes a 2036 multiple of about 14 times. The 11.2-times row is the case where the multiple falls to what my cash flows are worth at 8.5%. The 17.3-times row lets today’s multiple stay. That is the real bet in this price: not a faster business than my base, but a multiple that does not come down, for a business then growing 3.5% rather than 9%.

The discount rate

I discount at 8.5%. The capital asset pricing model with the 4.91% ten-year Treasury and Damodaran’s 4.41% implied equity risk premium gives 7.2% on Cencora’s beta against the market over 119 months, 7.5% over 60 and 5.9% over 36 C (calculated).¹⁴⁴⁷⁴⁶ The betas are low, 0.52, 0.58 and 0.23. I take a higher rate because a distributor with a quarter of its revenue in one customer and legal claims that run to 2039 carries more risk than its price moves show. 8.5% is also the rate on my Costco page.

At 7.5%, the top of the CAPM range, the base is $322; at 9.0% it is $213 E (estimate). Only the corner of the table with the lowest rate and the highest growth gets past the price. The price sits near my bull case, which agrees with the cases below.

Three cases

My base case has Cencora growing adjusted operating income 9.0% in fiscal 2027, helped by a full year of OneOncology and the first part of EyeSouth, then fading to 3.5% by 2036, an average of 5.9% against management’s target of 7% to 10% E (estimate).⁸ It gives credit for the weight of specialty, but only a 12% return on the money spent. That gives $8.5B of operating income in 2036.

The bull case has the top half of management’s range for five years, and a 16% return on new capital E (estimate). The bear case has a large customer repriced or lost around the 2029 renewal, taking six and four points off growth in two years, payment terms shift against the company, and the acquisitions earn 6% E (estimate). Even the bear is a going concern. It is only a poor purchase at $337.84.

The claims bridge

I leave MWI in the numbers rather than add it. The sale price is $3.5B, which is 5% of the market value, but MWI’s earnings are in my starting point, and the proceeds come as $1.25B in cash and $2.25B of preferred and common units in a private company that I can’t price.⁷

What I can’t judge

Some inputs I can’t know, and I would rather say so. I don’t know what the physician networks earn: the company does not give their profit, and the whole case turns on that, as the swing from $90 to $322 shows for a 6% to a 25% return. I don’t know what Sycamore wants from Walgreens, or whether the 2029 agreements will be extended, repriced or replaced. I don’t know how the opioid claims that are not accrued will end. I don’t know whether drug makers will send more drugs direct to patients. And I can’t know what multiple the market will pay in 2036. The first of these matters more than the rest, which is why I show it as a range and don’t pretend to a precise value.

Multiply it out

My base needs several things at once. Adjusted operating income growth near 6% a year on average: I give that 55%. The Walgreens agreements renewed on similar economics: 70%. The physician networks earning 12% or more on what is spent: 50%. No drug policy that cuts the distributor’s margin: 80%. Suppliers’ terms holding: 85%. Multiplied, that is 13% E (estimate). They overlap, since good growth helps the returns and the renewal, so the true figure is higher, perhaps a fifth. The price needs more than the base. It needs the base and a multiple near today’s in 2036. If I give that one chance in three, the joint odds are about 4% before overlap E (estimate). The argument for the price is not simple. It needs many things to go right together.

Quality and price are two questions

How good is the business? Steady and useful, with a margin that has held for ten years and a growing slice in specialty. Not a great business: its margin is thin, two customers are 38% of revenue, and it has spent part of the decade writing off what it bought. How good is the price? Not good on my numbers. At $337.84 it is 39% above my base value and just under my bull. A decent business can be a poor purchase at the wrong price, and on this evidence that is what this looks like.

It helps to compare it with the next best use of the same dollar. A ten-year Treasury pays 4.91% with no risk to the dollars promised.¹⁴ Cencora returns about 7.3% a year on my base, about two and a half points more, with all the risk of a share. On this site, Mastercard returns 8.4% a year on its base case, with a far wider margin. McKesson, which I did not value here, trades at 24.5 times last-12-month adjusted EPS against Cencora’s 19.7 C (calculated).³⁴¹² Against that, Cencora’s return is lower, and McKesson is priced at a higher multiple still.

The margin of safety

My margin-of-safety price is $170, 30% below my base value. At $170 my bear case returns 6.5% a year, above the Treasury, and the base returns 9.9% C (calculated). A 30% cushion is what I use for a business of this predictability. It would be wider for one I understood less well, and a 30% cushion here is not wide given that the biggest input, the return on the physician networks, is one I can’t see. The price is 99% above it.

VIIThe case for the price

What most people miss

First, the price is where it was a year ago: $337.81 at the end of October 2025 and $337.84 now, while expected adjusted EPS went from $16.00 to at least $17.90 C (calculated).¹⁴⁵ Second, suppliers fund $9.0B more than receivables and inventory, and the opioid settlement that cost $6.6B to accrue in 2020 is being paid at about $0.4B a year C (calculated).²¹⁷ Third, MWI is being sold at an enterprise value of $3.5B, about $18 a share, a price for profit that is already in the earnings C (calculated).⁷

The bull case

In plain words: Cencora is one of three national firms that move the country’s drugs, volume rises whatever the economy does, and it is adding the fastest-growing part of pharmacy, drugs given in a doctor’s office. The clearest statement is management’s own. In December 2025 it raised its long-term targets to adjusted operating income growth of 7% to 10% and adjusted EPS growth of 10% to 14%, after the OneOncology deal.⁸ IQVIA expects US drug sales to grow about 8.4% a year to 2029.² I found few named investors with a public case. The largest holders are index funds, Vanguard at 11.6% and BlackRock at 8.4%.¹³ Wellington Management added a little each quarter, to 4,559,761 shares at 30 June 2026.³⁷ Altimeter, Atreides, ARK, Akre, Fundsmith, Baupost and Berkshire Hathaway reported none at 30 June 2026, and Pershing Square none at 31 March.³⁷ The outside argument for the price is McKesson: at its 24.5 times, Cencora would trade near $420 C (calculated).³⁴¹²

Against reality

Much of the bull case is true today. The margin has held for ten years, the U.S. segment’s gross margin rose 52 basis points with the new networks, and guidance has been raised twice this year.⁶⁴⁵¹⁰ My bull case gives it all of that, with 11% operating income growth in 2027 fading to 5% and a 16% return on new capital. It is worth $353 E (estimate). So if Cencora does what management says and the networks pay well, it is worth about what it costs. To reach $337.84 I need the top of management’s range for ten years and a better return on the money, or a 2036 multiple as high as today’s. The gap is the return on new capital: the filings show $5.7B for RCA and $7.4B for OneOncology and $2.1B of physician claims, but not what the networks earn, so I use 12%.⁶

What a 10x would take. Ten times $337.84 is $3,378 a share, about $645B of market value, 5.5 times McKesson’s $116B C (calculated).³⁴³³ At 18.9 times earnings that needs about $34B of adjusted net income, 11 times fiscal 2025’s $3.1B, or about $3.5T of revenue at the 0.97% net margin, 11 times fiscal 2025 C (calculated).⁴ In ten years that is profit growth of 26% a year with no change in the multiple. The price rose 4.8 times in the last nine years, from a $15B company, helped by a multiple that rose 1.7 times C (calculated).¹ From $64B, tenfold is close to impossible for a distributor earning 1.3 cents on the dollar, and I would not build a case on it.

Where I come down

The supplier float and the MWI sale moved me: the cash is steadier and the balance sheet less stretched than the headlines suggest, and both are in my numbers. The management targets did not move me on value, since my bull case, which gives most of them, is worth about the price. What would raise my values most is the company showing what the networks earn, at 16% or more on the capital, and the Walgreens agreements extended beyond 2029 on the same terms.

VIIIRisks, and what would change my view

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

• Drug pricing policy. The 10-K describes an executive order on most-favoured-nation pricing, direct-to-consumer programs that may skip wholesale distribution, Medicare price negotiation under the Inflation Reduction Act and changes to the 340B program.² Any of them can shrink the manufacturers’ revenue the distributor’s fee is built on, or move volume around it.
• Brand price inflation and generic deflation. The company’s gross profit on brands and generics depends on how fast manufacturers raise prices and how fast generic prices fall, which it does not control, and a lower LIFO credit would be a cost.²
• Opioid claims not yet settled. The Fourth Circuit sent the West Virginia case back, a Department of Justice suit under the Controlled Substances Act is in discovery, and the company cannot estimate the range of loss for what is not accrued.⁶
• Regulators on the physician networks. The Federal Trade Commission has asked for more information on the MWI merger, and the company is buying physician-practice platforms in two specialties.⁶ The anti-kickback and False Claims Acts apply to both sides of those relationships.²
• International. Türkiye is highly inflationary, with $38M of expense in nine months and most of Alliance’s bank debt, and Alliance’s profit has been volatile.⁶ A 2024 cybersecurity event led to exfiltration of data.²
• Leverage and ratings. The company issued $3.0B of notes and drew $1.5B of term loans in 2026, and has committed another $2.8B for two deals, with interest rising from $292M to about $490M.⁴³¹⁰⁹⁵
• People. The CEO, chair and CFO have all changed in twenty months, and the finance chief started in June.³⁶

What would raise my base value: the fiscal 2026 results on 4 November showing fourth-quarter EPS at or above $4.59, fiscal 2027 guidance at the top of management’s 10% to 14% range, and a disclosure of what RCA and OneOncology earn. What would lower it: a falling gross margin in the U.S. segment, a larger charge for physician units, or any change in the Walgreens terms.⁵