Life after the empire

Life Healthcare spent the better part of a decade building a business outside South Africa: a hospital chain in India, another in Poland, a diagnostic-imaging group in Britain and a radiopharmaceutical developer in Europe. To fund the British deal alone it raised nine billion rand from shareholders in a single rights issue. Since 2019 it has sold every piece of it, handed the largest proceeds back as two special dividends, and returned to being what it started as: one of the two big private-hospital operators in southern Africa. The share price fell the whole way through that retreat, from about R24 in early 2022 to around R10.83 today. That looks like a company in freefall to all-time lows. It is not. Adjusted for the large dividends Life has paid, the shares bottomed at the end of 2021 and have handed cash back steadily since. Much of the nominal decline is capital returned, not value destroyed.

The market has read the shrinking revenue, the reported losses from the disposals and a visibly softer profit margin as a company in decline. We think it has misread three things. The reported losses were non-cash accounting entries tied to the exits, not operating failure. The margin that fell after Covid has not recovered even though the hospitals are as full as they were in 2019, which points to a fixable cost problem rather than a broken business. And the regulatory fear that hangs over the sector, National Health Insurance, is carried in equal measure by Netcare, whose shares have risen by more than two-thirds since the Act was signed while Life's kept falling. The same risk cannot explain two opposite outcomes.

On our numbers Life is worth about R14 a share through the cycle, in a range from roughly R11 in the bear case to about R20 in the bull. The bear case is close to today's price, so the buyer here is paying for the scenario in which the margin never recovers and getting the recovery, a strong balance sheet and a retained Alzheimer's-diagnostics earnout for very little. It is a bet on management finally delivering a margin it has guided to and its own peer has already achieved, with a genuine margin of safety underneath. Our rating is Buy.

Start with what Life Healthcare is today, because it is a different company from the one most investors remember. It runs 47 acute hospitals in South Africa, Botswana and Namibia, a little over 8,000 acute beds, and a set of adjacent businesses around them: renal dialysis, mental health, acute rehabilitation, oncology and diagnostic imaging. Nearly all of its money comes from patients covered by private medical schemes. It is a defensive, essential-service business protected by a licensing regime that makes new hospitals very hard to build. In the year to September 2025 the southern African operation turned over R25.1bn and generated about R3.8bn of normalised earnings before interest, tax, depreciation and amortisation.

The share price does not reflect any of that. Between 2016 and 2018 Life bought Alliance Medical and Life Molecular Imaging to go with hospitals it already owned in India and Poland, and for a few years it was a South African hospital operator with an offshore growth story bolted on. Then the story was dismantled, one country at a time, and here the share-price fall needs a second look. The drop from about R24 in early 2022 to R10.83 today looks like a collapse, but close to R9 a share of it was capital returned rather than value lost: the proceeds of selling Alliance Medical and Life Molecular Imaging were paid straight back to shareholders as two large special dividends. A company that hands back a third of its value in cash will show a falling share price whether or not anything is wrong with it.

There is a longer arc behind this. The raw share price has been falling since 2014, when it peaked near R44 and the market valued Life as an international growth story. Adjust the price for dividends and the picture turns over: on a total-return basis Life's low was not this year, nor the one it appeared to make in 2024, but the end of 2021, and the shares are worth more today than they were then. The raw chart shows a company falling to an all-time low. The dividend-adjusted chart shows one that bottomed four years ago, at the height of its complexity, and has handed a great deal of money back since. Measuring the fall from a 2014 peak the company should never have commanded, while ignoring close to R20 a share of dividends paid along the way, overstates the damage badly.

The question this report answers is whether the market has confused a company that got smaller with a company that got worse.

We think it has. Underneath the disposals sits a hospital business that is growing its revenue, running its wards as full as it did before the pandemic, earning a return on capital comfortably above its cost of capital, and sitting on one of the strongest balance sheets in the sector. What it is not doing is converting that activity into the profit margin it used to. That gap, between a healthy top line and a softer margin, is the whole debate, and most of it is a cost problem the company can address rather than a demand problem it cannot. At around R10.83 the shares trade close to our bear case; the downside is protected and the recovery comes cheap.

The core of Life is its South African acute hospitals, and the first thing to know about them is that they are hard to compete with. You cannot open a private hospital in South Africa by finding a building and hiring doctors. You need a licence, granted bed by bed through a provincial process that is slow, political and stingy with new capacity. The existing operators hold most of the licensed beds in the country, and that scarcity is the moat. It is the same kind of state-granted barrier that protects a fishing quota or a spectrum licence, and it is why private hospitals in South Africa have historically earned good, stable returns.

Around the acute hospitals Life has built a set of adjacent services, and this is where most of its recent growth has come from. Renal dialysis has expanded quickly, with treatment volumes rising from under 200,000 to over 500,000 in three years after Life bought a network of clinics. It runs mental-health facilities, acute rehabilitation units, oncology centres, and a growing diagnostic-imaging business assembled through small acquisitions, alongside Life Nkanyisa, a public-private partnership that provides mental and chronic care to state patients. These adjacencies are the growth engine now that the acute-bed count is broadly fixed by licensing — and, as the next section explains, they carry lower margins than the acute hospitals, which is part of why the group margin has drifted down.

The demand side is defensive but not growing. Roughly 90% of Life's revenue comes from patients on private medical schemes, and that pool has been broadly flat for years, held down by a weak economy and high unemployment. It sat at about 9.2 million people in 2024, a little over a sixth of the country, and grew by less than half a percent that year. That is the hard ceiling on the business. Life can do more for each insured patient, and charge more, but it cannot easily grow the number of them, so its growth has to come from the adjacent services and from price rather than from volume. It is a business to own for its cash generation and its defensiveness, not for volume growth.

And it does generate cash. In the year to September 2025 the southern African operation produced about R4.1bn of operating cash flow and spent roughly R1.4bn keeping its hospitals maintained, leaving well over R2bn of free cash before any expansion. Against a market value near R16bn, that is a free-cash-flow yield in the mid-teens, from an essential-service business with a licensing moat. The reason it is cheap is the margin, which is the next section.

Here is the single most important chart in the business. In 2019, before the pandemic, Life's southern African operation earned a normalised EBITDA margin of 23.8%. In 2020 Covid knocked it down to about 17%. Six years later, in 2025, it sits at 15.3%, lower than the Covid year itself. Revenue over that period grew by more than a third. The margin went the other way.

The obvious explanation is that the hospitals are emptier than they used to be, and for a while that was true. It is not true now. Occupancy fell from about 70% in 2019 to 58% at the depth of Covid, and by 2025 it was back to 69.7%, essentially exactly where it started. The wards are as full as they were before the pandemic. Yet the margin is eight points lower. Whatever is depressing profitability, it is not empty beds.

So what is it? Two things, and the split between them is the crux of the whole investment case.

The first is a squeeze between what Life charges and what it costs. Hospital tariffs are negotiated with medical schemes and have been rising at around 4% to 5% a year, while the cost of running a hospital, its nurses, agency staff and electricity, rises faster. Medical inflation in South Africa has run at about 8% a year for more than a decade, roughly three points above general inflation, and a hospital's wage and energy bills sit at the top of that range. The schemes on the other side of the table are wrestling with their own steep cost increases and push back hard on what they will pay hospitals. So Life's prices climb at 5% while its costs climb at 7% or 8%, and the margin narrows even when every bed is full. This is a real industry problem, and it is not unique to Life.

The second is a change in the mix of what Life does. The fastest-growing parts of the business, renal dialysis and the state care partnership, earn much thinner margins than a surgical ward. The acute hospitals run at around a 15% margin, complementary services higher, and the healthcare-services division under 10%. As the low-margin adjacencies grow faster than the acute core, the blended group margin drifts down even if nothing inside any single division gets worse. Some of the decline, in other words, is the arithmetic of a deliberate growth strategy, not deterioration.

Neither of those is a broken business. Management is running a cost-savings programme worth about R400m over three years and has set a target of getting the margin back above 17% by 2029. The bull case is that it does this: occupancy is full, the cost base gets tightened, and the margin claws back three or four points toward where it used to be and where its peer already trades. On R25bn of revenue, each point of margin is about R250m of profit, so the difference between a 15% and an 18% margin is around three-quarters of a billion rand of EBITDA. At the multiple this business trades on, that is most of the upside in the share.

The bear case has to be given its due. Life has now guided to margin recovery for three years and delivered the opposite. The mix shift toward lower-margin services is structural and continuing. And the pricing squeeze depends on schemes agreeing to tariff increases that keep up with cost inflation, which a flat insured population gives them every incentive to resist. A reasonable sceptic can conclude that 15% is the new normal and the 17% target is a number in a presentation. That is the debate, and it is not settled by anything in Life's own accounts. It is, however, largely settled by looking at the company next door.

Netcare is the other big private-hospital operator in South Africa. It runs the same kind of hospitals, serves the same medical schemes, negotiates the same tariffs, lives under the same regulatory cloud, and has the same September financial year. If there is a single clean control experiment available anywhere in this analysis, it is Netcare, and it answers two of the biggest questions in the bear case.

The first is whether the margin decline is a Life problem or an industry problem. The answer is both, and the proportion matters. Netcare leases most of its hospital property while Life owns most of its, and the accounting for leases flatters the margin of a company that rents. Put both on the same footing and the picture is clear: Life's margin fell about nine points from 2019 to 2025, Netcare's about five. So roughly five points of Life's decline is an industry-wide squeeze that hit the peer too, which the bull case has to respect. But the remaining four points is Life underperforming its direct competitor. In 2019 Life earned a higher margin than Netcare; today, like-for-like, it earns slightly less. It has handed its rival a lead it did not used to concede.

That four-point gap is the good news dressed as bad news, because a company-specific gap is one a company can close. It is exactly the kind of self-inflicted, fixable underperformance that a determined management with full hospitals and a cost programme can recover. And it comes with a caveat that runs in Life's favour: because Life owns its hospitals rather than renting them, it earns a higher return on the capital it employs than Netcare does, around 18% against 13%. Life is not a low-quality operator that has always trailed. It is a high-return business that has let an operating gap open up, and that is a more recoverable situation than the headline margin suggests.

The second question is the one that hangs over the whole sector: National Health Insurance. The government signed the NHI Act in 2024, setting out a state-run health fund that would, as written, sharply restrict the role of private medical schemes. Taken literally, it is an existential threat to private hospitals, and it is the single biggest reason the sector trades cheaply. But look at what the market has actually done with that fear. Netcare's shares bottomed at R11 in May 2024, the very month the Act was signed. Since then, as the Act became mired in constitutional challenges and it became clear that implementation is years away, unfunded and contested, Netcare has climbed back to over R18. The market priced the fear at the moment of signing and has been steadily taking it back out ever since.

Here is the point. Life carries exactly the same NHI risk as Netcare. Same law, same regulator, same exposure. Yet since the Act was signed the market has moved the two in opposite directions. As the fear receded it drove Netcare up by more than two-thirds, paying a distinctly higher multiple for the same kind of earnings. Life got the identical tailwind, and at first the identical bounce: through the back half of 2024 both shares re-rated hard as the NHI fear drained away, Life if anything leading. Then they parted. Netcare held its re-rating and pushed higher; Life handed its back, sliding through 2025 and 2026 as its own margin kept slipping. On a total-return basis Life is up only about a quarter over the two years, against Netcare's two-thirds, and even that quarter is mostly the special dividend it paid rather than a higher price for the business. On flat-to-lower prices and slightly higher earnings, Life's multiple has fallen while the peer's has climbed.

If NHI were what held Life down, the relief that lifted the peer would have lifted Life too. It did not, because a company-specific problem was pulling the other way. Same regulatory risk, opposite treatment: that contradiction is the strongest single piece of evidence in this report, and it says Life's problem is the company, not the law. What has actually sunk the shares is the margin gap and the ugly optics of the retreat, and both are more fixable than a change in the law.

To understand why the market is so sour on a business this defensive, go back to what Life did with its shareholders' money in the last decade, because the scars from that period are a large part of why the share is where it is. Life's ambition was to become an international healthcare group, and for a few years it was one. It built a hospital chain in India, bought Scanmed in Poland in 2014, acquired Alliance Medical, a British and European diagnostic-imaging company, for about £553m in 2016, funded with a R9bn rights issue that enlarged the share count by more than a third, and in 2018 added Life Molecular Imaging, whose main asset was a diagnostic agent for Alzheimer's disease. Four countries, four businesses, and a growth story the market was happy to pay for.

Then Life took it all apart, in order and without drama. It sold its Indian stake in 2019 for R3.7bn, at a profit of about R900m, and used the cash to cut debt. Scanmed followed in 2021, Alliance Medical in early 2024, and Life Molecular Imaging in 2025. A six-year, four-country retreat, each exit banked and the largest proceeds returned to shareholders rather than rolled into the next adventure. The discipline of the selling is a point in management's favour that the share price gives it no credit for.

Alliance Medical did not deliver what it promised. It grew its revenue well, by more than 60% over the hold, but its profit grew far less, by about a fifth, because it needed heavy and continuous capital investment to keep expanding. Measured properly, the British business earned a return on capital of around 14% or 15%, roughly what it cost to fund. What saved it was the exit: in early 2024 Life sold it to an infrastructure fund for an enterprise value of £910m, took about £593m of cash, and returned R8.8bn to shareholders as a special dividend, helped by a currency that had weakened by nearly 40% over the hold. The verdict is that Alliance Medical was competent capital recycling rather than value creation — a better outcome than most South African offshore adventures, but not the growth engine the market once paid up for, and its removal is one reason the multiple came down.

Life Molecular Imaging followed in 2025, sold to a US radiopharmaceutical company. The headline was ugly: Life booked a loss of about R2.4bn. But that was almost entirely a single accounting entry, the revaluation of an old profit-share liability owed to the business's original seller, which Life then settled in cash. The economics were better than the headline. Life took about R3.7bn of net proceeds up front, returned most of it as another special dividend, and kept a set of earnouts worth up to $400m running out to 2034, tied largely to the sales of the Alzheimer's diagnostic. Those retained earnouts sit on the balance sheet, valued conservatively, and the market ascribes almost nothing to them. They are a free option on one of the more interesting stories in diagnostics, thrown in with the shares.

It is worth putting the offshore chapter next to the one cautionary tale the sector already had. Netcare made its own move offshore a decade earlier. In 2006 it led a consortium that bought the largest private-hospital group in Britain for an enterprise value of £2.2bn, at the top of the credit cycle, loaded with debt and locked into 30-year property leases. Netcare's own share was a controlling stake, taken with a direct equity cheque of about £217m, the rest funded by its consortium partners and by debt piled onto the business itself. The British market turned, the leases became a noose, and by 2018 Netcare had written its whole investment down to nothing (the £217m of equity, plus the loans it later advanced to keep the business alive) and handed the debt-laden operation to its lenders. It did not lose the headline £2.2bn, which was the leveraged structure as a whole, but it lost every cent it had put in. Set against that, Life's round trip, out with about £553m and a rights issue, back with £593m and a special dividend, is a modest success. Both companies went to Britain. One lost its entire stake; the other came home roughly whole. Netcare was punished for BMI in its time: its shares more than halved between 2015 and 2020 as the UK mess unwound, and it has since moved past it. So the point is not that the market spared Netcare and singled out Life. It is that Life's offshore chapter, the thing its rating seems to hold against it, was the more disciplined of the two, and a company that recycled its foreign capital roughly whole has less to answer for than the share price implies.

At a glance the September 2025 balance sheet makes Life look almost debt-free: net debt of just R141m against R3.8bn of annual EBITDA. That impression needs one adjustment, because in the six months after that date Life settled the R2.4bn it owed the original seller of the radiopharmaceutical business. By the March 2026 half-year the balance sheet showed net debt closer to R3bn. That is the figure to use, and it is still conservative, a little under one year of EBITDA, against a business with the highest local credit rating and R1.8bn of undrawn facilities.

There is also hidden value in how Life is built. Unlike Netcare, which rents most of its hospitals, Life owns most of its property, and has been buying out leases to own even more. That property sits on the books at low historical cost, which is part of why Life earns such a high return on capital, and it is real asset backing under the share price. A hospital operator that owns its hospitals has a floor beneath it that a pure lease-based operator does not.

On capital allocation the record is good, with one fair criticism. Life has returned enormous sums to shareholders, R8.8bn and then another R3.4bn, both special dividends funded by the disposals, and pays a growing ordinary dividend. Its incentive scheme pays executives on headline earnings per share and return on capital rather than on revenue or size, which is what you want in a business that could otherwise be tempted to chase growth for its own sake. And there is a more pointed piece of evidence on how management reads the value: in June 2026, with the shares near R10.50, the chief executive bought a further R11.3m of stock on the open market with his own money. That is a discretionary purchase, not a scheme award, and about the clearest signal an insider can send. The picture is not uniformly one way: over the preceding months several executives were net sellers of mostly-vested stock, but the single most important insider, buying the most stock with his own cash near the low, is a genuine confirmation of the case.

The fair criticism is the special dividends themselves. With the share near its lows and below what we think the business is worth, buying back stock would have been more valuable to continuing shareholders than paying it all out. Life has done no meaningful buybacks. Its most senior executive plainly believes the shares are cheap, buying with his own money, yet the board has not followed with the company's. A buyback at these levels would be a strong statement and an accretive use of the cash.

A word on the governance record, since a diligent reader will check it. The business comes up clean, with two matters worth naming. The first is the one most South Africans associate with the name. In 2016 the Gauteng provincial government moved around 1,500 psychiatric patients out of Life Esidimeni, a state-funded facility Life operated, into cheaper and largely unlicensed non-profit homes, and at least 144 of them died of starvation, neglect and exposure. It was one of the gravest human-rights failures of post-apartheid South Africa. Responsibility, as the Health Ombud and the arbitration that followed both established, lay with the provincial government, which forced the transfer over Life's objections. Life was the provider the patients were taken from, not the cause of what happened to them. The association is unfair, but because the company's name sits on the tragedy it is worth saying plainly where the fault lay. The second matter is far smaller: a 2024 media report questioned Life's funding of a state mental-health non-profit it services, which the company rejected as factually inaccurate, noting the contract had run eight years with clean audits and that the government department itself owed the non-profit R172.6m for work already done. In neither case was Life the party at fault. On its own conduct, the record is clean.

Life is close to a single business now, a South African hospital operator, so it does not need the divisional break-up a conglomerate would. We value the core hospital business on its earnings, add the pieces that sit outside those earnings, subtract the debt, and check the answer against the ordinary measures. It earns about R3.8bn of normalised EBITDA today, at a depressed 15% margin. The single biggest judgement is what margin to use through the cycle. We do not assume a return to the old 24%, because even the healthy peer sits well below that now; we assume the margin recovers part of the way, closing some of the gap to Netcare, and run it as a range.

The base case comes out at about R14 a share. Two judgements move it more than anything: the margin, which is the whole thesis, and the multiple. On the multiple we have deliberately used a small discount to the peer, 5.5 times against Netcare's roughly 6. Life's stronger balance sheet and higher return on capital would ordinarily argue for parity or a premium, but its margin is falling while Netcare's rises, and a market that pays less for the operator with the worse trajectory is behaving rationally. On normalised earnings of about 107 cents a share, R14 is roughly 13 times earnings, a shade below where Netcare trades. At the current price near R10.83, Life trades on about 10 times, a clear discount to the peer for a company with a stronger balance sheet. A discounted cash flow lands in the same place, about R14. Every method points to the same conclusion: the shares are worth somewhere in the low-to-mid teens, comfortably above R11.

On the discount rate, and where we differ from the market

It would be easy to assume the gap between our R14 and the market's R11 is the discount rate. It is not. Our cost of capital is built from live data: an 8.4% risk-free rate off the South African ten-year bond, a South African equity risk premium of about 8.5% (a mature-market premium plus the country risk premium the sovereign spread implies), and a beta of 0.68 measured off the stock itself. That comes to roughly 13%, almost exactly the WACC Life discloses for its own investment decisions. The external rates nearer 16% imply a beta close to 1.0 that no regression on the share supports and that the company's own figure contradicts; they are the output of models that floor beta at 0.8 by rule and apply it in bulk, not a considered view of this business.

So the rate is not the disagreement. At around 13%, on cash flows that reflect a business whose insured pool cannot grow, where revenue is essentially price plus a modest margin recovery fading to a low terminal, the DCF lands at about R14, the same as the relative value. Where we differ from the market is the cash flows, specifically the odds of the margin recovering. The price near R11 is the market pricing the bear case, that the margin never comes back; we weight a partial recovery more heavily, because the peer proves that margin is achievable and occupancy has already returned. That is a probability judgement, not a claim the market has made an arithmetic error, and it is the one thing on which the whole call turns. We hold the DCF loosely, as corroboration of the low-to-mid teens rather than a precise figure, because its answer swings widely with growth and terminal assumptions no outside analyst can pin down. The relative value and the probability-weighted scenarios are the anchor.

The verdict

At around R10.83 Life trades close to our bear case, the scenario in which the margin never recovers at all. The buyer today is paying for permanent stagnation and getting, for almost nothing, the possibility of a margin recovery that management has guided to and its own peer has already delivered, plus a strong balance sheet, real property backing and a free Alzheimer's earnout. The downside is protected by the price and the assets; the upside is a specific, nameable event: Life closing the operating gap to Netcare.

This is a bet on execution, and the bet is not free of doubt. Life has guided to margin recovery three years running and delivered decline. Part of the pressure on its margin is a real industry squeeze that will not lift. And the dividends, welcome as they are, are not the buyback that would signal real conviction from the board. So the framing is this. Life Healthcare is a good, defensive, cash-generative business with a licensing moat, a fortress balance sheet and hidden property value, trading near its lows because the market has confused a company that got smaller with a company that got worse. The margin of safety is real. The catalyst is identifiable. What is required is for management to finally do the thing it has promised. On that basis, and at this price, we think the odds favour the patient buyer. Our rating is Buy.