Selling HL after investors accept a low-ball takeover offer

Of all the companies I’ve invested in over the last few years, there’s a good chance that Hargreaves Lansdown (HL) was the best.

It was started by Peter Hargreaves and Stephen Lansdown in 1981, with the pair initially operating out of Hargreaves’ spare bedroom, and it has been a spectacular success almost from day one. 

The initial idea was to sell unit trusts through the post, with clients acquired through newspaper adverts. This was by no means a novel approach, but Hargreaves’ marketing genius and desire to build the biggest financial advice firm in the UK, along with Lansdown’s desire to build the best financial advice firm in the UK, combined to create an unstoppable business that achieved both of their goals.

Today, HL has a crushingly dominant market share of more than 40%, with returns on capital that other companies can only dream of. 

However, despite that success, shareholders recently accepted a low-ball takeover offer (of £11.10 per share) from a private equity consortium, and HL is now expected to be taken private in the first quarter of 2025.

As a shareholder since 2020, I find this deeply disappointing, especially as the UK stock market is about to lose another exceptional business.

I could have held onto my shares until they were prised, unwillingly, from my grasp next year, but I have decided to cut my losses now and reinvest the proceeds elsewhere.

As usual, this post-sale review will cover why I invested in HL in the first place, what happened while it was in the portfolio and what I think of the company’s valuation today.

Why was Hargreaves Lansdown added to the portfolio?

• Hargreaves Lansdown purchase review: August 2020 (PDF)

You can read my full pre-purchase review above, but the short version is that I added HL to the UK Dividend Stocks Portfolio because it was an almost perfect example of a high-quality progressive dividend stock and its results at the time (shown below) were virtually peerless.

The most striking feature of HL’s performance is its almost perfectly straight growth trajectory, fuelled by an unbelievably high net return on capital that averaged almost 60% per year over ten years (most companies struggle to reach 10%). On top of that, it had no debt, hadn’t made any acquisitions and had little need for cash-draining capital expenses.

The only red flag came from its high PE10 ratio (the ratio of share price to 10-year average EPS), which at almost 51 was far above the limit of 30 I used at the time. Unfortunately, I ignored that red flag and I’ll explain why shortly. But first, I want to explain why HL was such a good example of a quality dividend stock.

A capital-light business run exceptionally well by two genius founders

How did HL manage to generate a net return on capital of more than 50% for more than a decade?

The first half of the answer to that question comes from the nature of HL’s business. First and foremost, it’s a professional services firm offering stockbroking, financial advice and other financial services. To do that it needs people, phones, desks, computers and offices to put them all in, but what it doesn’t need is factories, warehouses, machinery, inventory or any of the other expensive physical assets required by companies that produce goods rather than services.

As the company grew, from selling unit trusts through the post in the 1980s, to building one of the early online fund supermarket platforms in the 1990s, to becoming the Waitrose of retail investing in the 2000s, HL has, of course, had to invest in offices, computers and so on. But, fundamentally, it is and has always been a capital-light people business and, increasingly, a capital-light technology business.

On the other hand, having a capital-light business model is a double-edged sword. Yes, it makes it easier to generate extremely high returns on the small amount of capital required by the business, at least for a while, but it also makes it easier for others to launch competing firms precisely because they don’t have to raise vast sums to buy or lease factories, machinery and other capital-intensive assets.

If a capital-light business generates high returns on capital but doesn’t have a durable competitive advantage, then a swarm of copycat firms will quickly appear and that profit pool will end up being spread, much more thinly, across many firms rather than one, so:

How has HL managed to protect such high returns on capital from the relentless onslaught of profit-hungry competitors?

The answer, at least from 1981 until 2010, is that the business had a durable competitive advantage in the form of its two genius founders.

"Genius founder" means different things to different people, but for me, it means someone who has built and run a business using many of Jim Collins’ concepts, described in his three main books, Built to Last, Good to Great and Great by Choice. In summary, HL’s founders applied the following concepts:

• Level 5 Leaders: Hargreaves and Lansdown were ambitious first and foremost for the firm, rather than for themselves (although they still became extremely rich)
• Hedgehog Concept: They had a deep understanding of what the firm could be the best at
• Preserve the Core & Stimulate Progress: They kept HL’s core business largely unchanged for decades, which built deep institutional expertise, but they were also willing to evolve the business as the environment evolved
• Fire Bullets then Cannonballs: They stimulated progress cautiously, by trying out new ideas on a small scale before pushing the accelerator on those that succeeded
• Productive Paranoia: They were paranoid about risk and refused to use debt or make large acquisitions

Eventually, HL’s advantage of having genius founders led to other competitive advantages, including its current market-dominating scale. That scale gives it economies of scale, but more importantly, it also gives HL a mild form of network effect. More specifically, HL can attract retail investors cheaply by offering them access to the widest range of investments, and it can attract investment providers cheaply by offering them access to the largest pool of investors.

These follow-on advantages are essential because founders can’t run their companies forever, so although having a genius founder is a competitive advantage, it isn’t a durable advantage. For a founder’s genius to become a durable competitive advantage, it must create other advantages that are durable, and one way for a founder to turn their genius into an enduring competitive advantage is to embed their genius deep within the company’s culture and processes.

To put it another way, there are two types of genius.

The first is the genius with a thousand helpers. These are people with towering capabilities who drive businesses to achieve seemingly impossible feats, but when the genius is no longer around, those businesses often drift into mediocrity.

The second type is the genius who wants their business to be even more successful after they’ve left than it was during their tenure. These leaders spend a lot of time embedding their genius into the company’s culture and processes. Just as importantly, they spend a lot of time on succession planning, ensuring the business can develop homegrown CEOs with deep expertise in the business and a burning desire to position it to succeed for decades to come.

Of the companies I'm currently invested in, PageGroup is probably the gold standard for this type of advantage.

Michael Page co-founded PageGroup in 1976 and since then it’s had just four CEOs. All of them were either Michael Page himself or insiders who joined at the bottom as recruitment consultants and meritocratically worked their way to the top. Other than the founder, each of PageGroup's CEOs had already worked in the firm for more than ten years before being offered the top job.

Although home-grow CEOs aren’t a silver bullet, they often provide a degree of stability, expertise and long-termism that can be a powerful and durable competitive advantage.

Unfortunately for HL, its founders didn’t build that kind of culture and, since their departure in 2009/2010, HL has had a series of outsider CEOs with little or no experience within the business.

With outsider CEOs who have little emotional connection to the business, there’s an elevated risk that they’ll take the business in the wrong direction or seek to boost short-term profits by underinvesting in long-term growth. In HL's case, it seems that both may have occurred after the founders left, and that’s partly why it’s being taken private.

Having said all of that, hiring outsider CEOs hasn’t been a total disaster for HL and when I added it to the portfolio in 2020, it was still a great example of a quality dividend stock.

So, if HL is such a great business, why has my investment produced an annualised return of minus 2.7%?

The answer is that I made one big mistake.

My one big mistake: Paying a high price for an exceptional business

When I reviewed HL in 2020, its share price was £17.17 and it had the following valuation stats:

• Dividend yield of 2.0% (my minimum acceptable yield at the time was 2%)
• PE10 of 51 (my maximum acceptable PE10 was 30)
• PD10 of 59 (my maximum acceptable PD10 was 60)

Note: PE10 and PD10 are the ratios of share price to ten-year average EPS and DPS respectively

Each of my three core valuation metrics was either at or above the self-imposed limits I used at the time, so why did I ignore these clear signs that HL was probably expensive rather than cheap?

The answer is that during 2020, I had started to move away from those metrics because I felt they were too focused on the past and not focused enough on the future. When it comes to investing, the future is what matters, so I’d started to value stocks based on a combination of the above ratios and a simple dividend model known as the Gordon Growth Model (GGM).

The GGM (also known as the yield-plus-growth model) is based on the idea that long-term returns are dominated by the combination of dividend yield and dividend growth, so the model calculates a stock's expected total return by adding its dividend yield to its estimated long-term dividend growth rate.

At £17.17 per share in 2020, HL had a dividend yield of 2%. HL also paid fairly regular special dividends and, on average, they added 0.4% to its yield. So, the estimated "total" dividend yield for HL was 2.4%.

Back in 2020, I used a target annualised total return of 10%, so for HL to look attractive I had to expect its dividend to grow by at least 7.6% per year (because 2.4% + 7.6% = 10%). So that is exactly what I did, and that was my big mistake.

My rationale for assuming HL could grow by 7-8% per year was as follows:

1. HL had grown by 12% per year over the previous decade, without using debt, so a growth rate of 7-8% was clearly possible from a historical point of view
2. HL was expanding beyond stocks and shares and into cash savings, wealth management and workplace pensions, so the addressable market seemed to be more than large enough to support high-single-digit growth for many years to come

However, what I didn’t fully appreciate at the time is that with a simple yield-plus-growth model, what you’re essentially assuming is that the company will grow at your specified growth rate, not just next year or for the next decade, but forever. Of course, for the vast majority of companies, assuming any kind of eternal growth rate is extraordinarily optimistic.

I stopped using these simple yield-plus-growth models in 2021 when I realised just how unrealistic they were, and I replaced them with a significantly more realistic dividend model which assumes the shares are sold after ten years, leaving the tricky question of what happens between now and the end of time to someone else.

My approach to these models has evolved over the last four years, so I think it would be useful to value HL, as it was in 2020, with the type of model I use today.

Here are the four numbers needed to build that model, using HL's figures from 2020:

1. Capital per share of 87p
2. 10-year average net return on capital of 58%
3. 10-year average dividend cover (including special dividends) of 1.3
4. 10-year average PE ratio of 33

And here’s the initial model, based purely on the above historical figures.

This model has HL’s dividend growing from 40p in 2020 to 100p in 2028, which requires an annualised dividend growth rate of 12.2%. And so, in the medium term at least, this model is even more optimistic about HL’s growth than my simple yield-plus-growth model, which used an expected growth rate of 7.6%.

However, instead of assuming that HL can continue growing forever, this model assumes the shares are sold after ten years at a historically average PE ratio of 33. With estimated 2028 earnings of 127p per share, that produces a sale price of £41.90, compared to the real-world share price in 2020 of £17.17.

Discounting the cash flows (the dividends plus the sale proceeds) back to 2020 at 7% per year (a fair market-average return for UK stocks) gives HL a fair value, in 2020, of £25.40 per share. The actual share price was £17.17, so this model is saying that HL was trading at a 32% discount to fair value.

I generally prefer a discount of at least 33%, but 32% is easily close enough for an exceptional business like HL. On that basis, my current approach to valuing stocks still seems to suggest that HL was attractively valued in 2020.

But that isn’t the end of the story. Having built many dividend models over the last five years, I know they can be optimistic, so I have a rule:

• Rule of thumb: If the forward dividend yield is below 3% when the shares are trading at fair value, your fair value estimate is too high

In this case, if HL’s shares were trading at the fair value estimate of £25.40, the model’s forward dividend yield (based on the model’s 2020 dividend of 40p) would be 1.6% ("Yield at FV" in the table above), so the model's fair value estimate is too high.

To fix this, I can tweak the model until the forward yield is 3% when the share price is at fair value. The easiest way to do that is to raise the dividend by reducing dividend cover. This effectively assumes management is unable to grow the business by 12% per year, so less cash is needed to fund growth, which means more cash can be paid out as dividends.

Here’s an updated version of the model with dividend cover lowered from 1.3 to 1.1.

Lowering the model's dividend cover increases the 2020 dividend from 40p to 46p and paying those higher dividends reduced the growth rate from 12.2% to a more achievable 5.3%. I also reduced the exit PE from 33 to 30, which is reasonable as lower growth usually leads to a lower PE ratio. Those changes did increase the fair value yield to 3% by raising the dividend and lowering the fair value estimate.

The updated model’s fair value estimate is considerably lower than the previous estimate, at £15.26 versus £25.40. In fact, this lower fair value estimate is below the 2020 share price of £17.17 and that means, according to the updated model, that HL’s share price was above fair value when I reviewed it in 2020 (it was 13% above, to be precise).

To reach my target discount of at least 33%, the share price would have to have fallen below £10.17, and it didn't do that until 2022.

On that basis, I’m confident that my current approach to valuing stocks would have kept HL out of the portfolio until at least 2022, by which point the share price had fallen to £10. Tellingly, if I'd purchased HL at £10 in 2022, it would have produced a very solid 9.1% total annual return to today.

However, the reality is that I did add HL to the portfolio in 2020 when it was trading above what I would now consider a conservative estimate of fair value. In fact, I ended up paying £18.40 per share rather than the review price of £17.17, as the share price jumped up just before I purchased the shares. And so, having unknowingly overpaid for an outstanding business, what happened next?

What happened during the holding period (2020 - 2024)?

Data source: SharePad

From 2020 to 2024, HL’s business continued to perform extremely well, with revenues growing by 59%, earnings by 19% and the dividend by 28%. All of this came during a period of extreme disruption, which is a testament to the stability and enduring strength of HL’s core business.

Although the company’s results were, in my opinion, impressive, its low-single-digit growth rate from 2020 to 2024 was far below the double-digit growth of the previous decade, and some shareholders (notably Peter Hargreaves) weren’t happy with the way HL was being run.

There was much wailing and gnashing of teeth in 2022 when the (outsider) CEO announced an updated strategy to pursue wealth management as an extension to the existing fund supermarket business model. Peter Hargreaves complained loudly and the CEO and his strategy were soon gone.

That was unfortunate and, in my opinion, having one of the founders throw rocks at the CEO certainly doesn’t help. Ultimately, this was much ado about nothing as the wealth management strategy was hardly a seismic shift from HL’s existing business model, which already included an element of personal financial advice.

Either way, HL’s incredible business continued to march forward, growing the dividend through the pandemic, albeit more slowly than before and with lower net returns on capital than before (although still exceptionally high at around 40%, versus 10% for most companies).

In addition to the concerns around HL’s attempted move into wealth management, there were growing concerns that HL’s high returns on capital had only been possible because it had consistently underinvested in technology. Underinvesting in the core business can be a great way to boost short-term profits and dividends, but it can also be a great way to undermine a company’s longer-term prospects, and it’s something you’re much more likely to see with an outsider CEO than a homegrown CEO.

With back office and customer-facing technology that was far from cutting edge, smaller, younger, “fintech” competitors were attracting the next generation of retail investors away from HL, and that was undermining the foundations of HL’s future growth.

To be fair to the previous CEO, his 2022 strategy did focus heavily on improving the company’s technology, but apparently it wasn’t enough and he was replaced, in 2023, with another outsider CEO who was hired specifically because of his technology background. It now seems that HL’s medium-term future will be dominated by an expensive technology improvement programme.

A combination of slowing growth rates, the pandemic, concerns around technology underinvestment and Peter Hargreaves’ habit of publicly throwing rocks at HL’s CEOs, all combined to drive the company’s share price down from a high of £24 in 2019 to a low of £7 in early 2024.

Fortunately, that lower share price allowed me to top up my investment at close to £8 per share, which at least partially offset my initial error of paying too much in 2020.

HL was clearly cheap at £7 and that's why a private equity consortium was more than happy to offer a mere £11.10 per share to take the company private.

That was a significant premium to the then-current share price of £7.50, but it was still less than half the company's peak share price. To my surprise, the board accepted the offer. The deal was put to shareholders in October and most of them agreed with the board, so HL will now almost certainly be taken private in the first quarter of 2025.

Why is the company being sold now? 

Unlike most of HL’s shareholders, I voted against the takeover because I don’t think £11.10 is an attractive exit price. In fact, my current dividend model has HL’s fair value closer to £17 than £11.

The world has moved on from 2020, so my current dividend model looks different to the 2020 model shown above, but the 2024 model has underlying assumptions (listed below) that aren’t hugely different:

1. In 2025, capital per share is 150p (double the level of five years ago as HL has invested heavily in upgrading its technology assets)
2. Net return on capital remains steady at 40% (lower than the historical average due to the significant step-up in technology investment)
3. Dividend cover is held at 1.2 (retaining enough earnings to fund a near-7% medium-term growth rate)
4. The shares are sold in 2033 at a PE of 25 (slightly below the average PE of the last ten years)

Those assumptions produce a model where the forward yield at fair value meets my minimum 3% requirement, so I’m confident that it’s both realistic and conservative (although, of course, that doesn’t mean it’s correct, because such exacting foresight doesn’t exist in the real world).

The fair value estimate comes in at £16.93, and at the start of November the share price was sitting at £10.88, so the discount to fair value was a healthy 36%.

If the share price suddenly jumped up to the takeover price of £11.10, the discount to fair value would still be good at 34%. That’s better than my 33% minimum discount, so the takeover price is one where I would be happy to buy (as, I'm sure, is the private equity consortium), so I'm definitely not happy to sell at this price.

Unfortunately, other shareholders voted in favour of the takeover, so it’s now virtually certain that HL will be taken private in the first quarter of 2025, at which point the UK stock market will lose one of its most outstanding businesses.

With the shares trading just 2% below the takeover price, I see little point in hanging on until the bitter end. I would rather remove HL from the portfolio now so I can get on with the more positive task of reinvesting the proceeds into existing and future holdings, and that is exactly what I've done.

I sold HL in early November and here are the final results:

• Holding period: 4 years, 3 months (August 2020 to November 2024)
• Capital gain: -17.8%
• Dividend income: 9.9%
• Total return: -7.9%
• Annualised total return: -2.7%

As for HL’s future, it’s an exceptional business and I’m also a client, so I wish it the very best of luck. Hopefully its private equity owners will improve the business by stepping up its investment in technology, and perhaps it will reappear on the UK stock market within the next five or ten years.

Unfortunately, like most private equity takeovers, the acquisition is being funded partly with debt; £1.75 billion to be precise. If that ends up on HL’s balance sheet then the company will immediately become highly leveraged, with a Debt Ratio (debt to ten-year average earnings) of 7.1, which is comfortably above my limit of six times. 

It would be a shame to see HL struggle under the burden of excessive debt, but alas, that has been the fate of many businesses after a period of private equity ownership.

On a more positive note, the sale of HL left the UK Dividend Stocks Portfolio with a healthy surplus of cash and that will fuel the purchase of two new holdings in the next few months, which will bring the number of holdings back to my target of 25.

I’ll end this review with a 1963 quote from the master of sensible investing, Ben Graham, which neatly sums up the error that led me to overpay for HL:

“We have been trying to point out that this concept of an indefinitely favorable future is dangerous, even if it is true; because even if it is true you can easily overvalue the security, since you make it worth anything you want it to be worth. Beyond this, it is particularly dangerous too, because sometimes your ideas of the future turn out to be wrong. Then you have paid an awful lot for a future that isn't there. Your position then is pretty bad.”