Depending on which guru you ask, diversification is either the only free lunch in investing – or something that only makes sense if you don’t know what you are doing.

The first quote is commonly attributed to Harry M Markowitz, Nobel Prize winner and father of model portfolio theory. The second idea is from Warren Buffett, who called diversification “protection against ignorance”. So who is right?

The answer is both, potentially. Markowitz argues that you can use diversification to achieve better risk-adjusted returns – a better balance between performance and volatility. As for Buffett, another way to look at his argument is that unless you know what you are doing to the extent that he does, you probably need to diversify.

Experts generally agree that for the vast majority of investors, putting all your eggs in one basket is a bad idea. But what does diversification actually look like, and how do you make it work for you?

The basics: time horizon, correlation and risk

There are degrees of diversification. Rather than diversify for the sake of it, you need to consider the idea in light of your time horizon, risk appetite and investment goals.

In the context of a multi-asset portfolio, diversification is typically seen through the lens of correlation, a measure comparing the behaviour of two assets over a certain period of time. A positive correlation means the two tend to go up and down in tandem. Multi-asset investors instead look for assets that, ideally, have a negative correlation with equities: they go up when stocks go down, cushioning the impact of crashes and bear markets.

Not every investor needs this. If you have a decades-long horizon as well as the risk appetite and sangfroid to sit tight during the ups and downs, you can afford to stay fully invested in stocks. Perhaps with one eye on this kind of time horizon and risk appetite, Buffett has argued that the average investor only needs a straightforward 90/10 portfolio – 90 per cent in an equity tracker, and the rest in short-term bonds.

So first, ask yourself what kind of investor you are, and what risk means for you. Technically speaking, risk is often measured through standard deviation – a statistical measure that looks at how much a variable (for example the price of an asset) deviates from its mean over a certain period of time. It is basically a measure of volatility.

But if you ask financial planners and investment managers about risk, they’ll often talk about the risk of losing their clients’ money, or the risk of not meeting their clients’ financial goals. When you consider risk this way, volatility suddenly becomes a lot less important.

Being a long-term investor with a high risk tolerance does not mean that you don’t have to consider diversification, however.

You may not need to worry about how your portfolio will behave in the event of a stock market crash tomorrow, but you need to consider that the fortunes of different countries, investment styles and sectors could change over the long term. Nobody can predict with any degree of certainty whether the US market or the Asian market will outperform the other over the next 20 years. So, you buy both US and Asian stocks, and avoid disappointment.

At the opposite end of the spectrum is an investor with a shorter time horizon, for example because they need to regularly draw money from their portfolio, or because they are risk averse. The short-term correlation between assets is a lot more important here.

This is the type of investor who might find it useful to understand how an investment trust such as Ruffer Investment Company (RICA) operates. The trust is focused on wealth preservation, so has two basic goals: not to lose money in any 12-month period, and to return “meaningfully” more than cash. A year is a very short time in investing, so its managers spend a lot of time on the hunt for uncorrelated assets.

Ruffer fund manager Ian Rees explains: “For one side of the portfolio, we’ll say: ‘which are the growth assets that will make us money in the good times?’ And typically, that’s going to be equities . . . then on the other side, we think: ‘what could go wrong? Are there any immediate risks that we see or worry about, whether in markets or in the economy?’ And connected to that: ‘how would we protect the growth assets if they fell?’”

“We would always expect parts of the portfolio to be making money, and also parts [of the portfolio] to be losing [money],” he adds.

Scott Robertson, investment specialist at St James’s Place, says diversification is a little like starting a business that sells both ice cream and umbrellas. “When it’s sunny, happy days, your ice cream is going to be selling, and when it is chucking it down, your umbrellas are going to be selling,” he explains. However, doing the same with investing is easier said than done.

In the IC’s Alpha section, you can find a range of multi-asset portfolios suitable for different investors and risk profiles, focused on achieving the best possible risk-adjusted returns.

Trust in shares and ditch the dollar (for now)

Diversified asset portfolio for 2026

How to think about asset allocation

Market regimes and diversification

An Alpha subscription gives you access to our most advanced features and content including more in-depth ideas, research reports and analysis. You can subscribe or upgrade here

What makes an uncorrelated asset?

If you are aiming for true diversification, adding a lot of different positions to your portfolio is unlikely to do the trick on its own. If you don’t pick carefully, there’s a decent chance that many of those positions will actually behave in the same way.

In doing so, you may end up with the exact opposite of what Markowitz was advocating for. ‘Diworsification’, a term coined by fund manager Peter Lynch, can apply to portfolio construction as well as stockpicking: by adding too many positions, or the wrong ones, you can actually add risk to your portfolio and/or worsen its returns. A classic example is holding so many active funds doing different things that you find yourself owning the entire market – as a single tracker fund would – but paying active management fees for the privilege.

On top of this, when there’s a shock to the system, assets do not always behave as you expect. “The interesting part, but also the difficult part, is that correlations change through time,” Rees says. A recent example was 2022, when equities and bonds fell at the same time, raising serious questions about the rationale behind the traditional ‘60/40’ portfolio. This is a common balanced portfolio, with a 60 per cent exposure to stocks and the rest in bonds.

David Storm, chief investment officer at RBC Wealth Management, says he thinks about diversification in terms of assets whose returns are driven by different factors. “Most people have portfolios that are dominated by two basic factors: interest rate risk and equity risk,” Storm says. If you want diversification that works in a broader range of scenarios, you might need something more sophisticated.

“For example, we like [betting for and against] commodities because commodities are, in the main, driven by something quite different from just GDP growth and central bank policy. They might be driven by . . . the weather in Texas, which is quite a diversifier for a portfolio,” he says.

Stocks: all about the US?

Diversifying an equity portfolio is fairly straightforward, especially if you use funds. A passive global tracker is a decent starting point, as it gives you access to hundreds of different companies in one go, operating in different sectors and geographies.

However, at present, global trackers are very exposed to US technology stocks, meaning that they are arguably not as diversified as they used to be. It can make sense to add more diversification with dedicated exposure to other regions, sectors and styles.

Giulio Renzi-Ricci, head of asset allocation at Vanguard Europe, lists the many reasons why this seems sensible. As a starting point, other regions and sectors offer cheaper valuations. The Magnificent Seven tech stocks also have a number of risks to contend with: artificial intelligence (AI) might not deliver all the disruption it has promised, or not do so fast enough; capital expenditure might not translate into significant enough revenues; current winners might be replaced by other players. Finally, he thinks AI could prove disruptive to a range of other sectors, including healthcare and financials, where its impact has not yet been fully priced in.

When it comes to equities, diversification does not necessarily mean negative correlation – a market crash in the US would most likely affect the rest of the world, particularly in the short term. When the US sneezes, the world catches a cold, as the saying goes.

But correlation doesn’t need to be negative for an asset to offer some degree of protection and diversification; a low positive correlation can still help cushion the blow of a crash. And when it comes to correlation between US stocks and the rest of the world, things might be shifting a little.

Dina Ting, global head of index portfolio management at Franklin Templeton, notes that the relationship between certain Asian markets (particularly Japan, South Korea and China) and US equities has been positive but very low over the past three years, and was even lower over the course of 2025.

“To us, this reflects shifting supply chains, trade policy uncertainty and country-specific growth drivers rather than broad US equity momentum,” she says. “That said, short-term investor sentiment can still overwhelm fundamentals. Tech-led risk-off episodes have triggered cross-border selling in Asia. In November, for example, we saw the largest monthly foreign outflows in nearly six years, following AI and tech valuation scepticism.”

She lists India and south-east Asia as the most insulated markets from a potential AI-related shock. Taiwan is likely to be the most impacted, while South Korea would also suffer due to its heavy exposure to chips and AI hardware (but exposure to shipbuilding and heavy industries would help). Japan is “middle of the pack”, she says, although the Nikkei benchmark is now more heavily weighted to AI plays than it once was. “Moves in US AI stocks can transmit through semiconductor equipment makers and electronics exporters, but Japan’s broader equity market is more diversified than Taiwan or Korea,” Ting says.

Are bonds actually back?

There are other correlation issues to consider. The 60/40 portfolio was built around the idea that bonds are uncorrelated to stocks. But since 2022, experts are divided on whether this still applies, and to what extent bonds still work as diversifiers.

The argument on the one hand asserts that the problem was artificially low interest rates – when bond yields hovered near zero, there really was just one direction for prices (which move inversely to yields) to move, and that direction was down.

Now that base rates have broadly normalised and yields are closer to more conventional levels, we are no longer likely to see bonds and equities meaningfully fall at the same time. If the stock market crashes, this argument says, we should see a classic flight to the safety of fixed income.

“You will have some protection from using fixed interest, if for example a more recessionary outlook unfolds [in 2026] and interest rates fall faster than expected,” says Rob Morgan, chief investment analyst at Charles Stanley.

You don’t need anything other than this to diversify your portfolio, adds SJP’s Robertson. “We’re fine having just equities and bonds because bonds are back, bonds are doing what they should be doing,” he says.

Others are not quite as confident, their main concern being inflation. Inflation is bad news for bonds, both because it erodes their real-term returns, and because it can raise interest rate expectations, pushing bond prices down.

“If we go from a disinflationary world to a more inflationary world, we don’t think you can rely on the bond/equity correlation being negative,” says Ruffer’s Rees. While he agrees that 2022 constituted an especially big shock, he expects inflation and the macroeconomy to continue to be much more volatile than they used to be, making for a bumpy ride for fixed income.

Renzi-Ricci argues that the correlation between equities and bonds has been higher than usual since the Covid-19 pandemic. Covid, he explains, was a supply-side shock to the economy, rather than your standard demand-side shock, and prompted central banks to take a more ‘proactive’ approach to monetary policy in order to stimulate the economy. In turn, this created a situation that combined both economic fears (bad for equities), and high inflation, which was then followed by aggressive base rate hikes (bad for bonds).

But the level of correlation between equities and bonds has been decreasing, and he expects this to continue as more conventional economic conditions are restored. Shocks can happen, he says, but that doesn’t necessarily imply that we now permanently live in an environment where equities and bonds are correlated.

However, he acknowledges that more surprises of this kind, which are usually caused by the likes of natural disasters or wars, are certainly not impossible. If you envisage them becoming more frequent, it will take more time to restore normalcy, and bonds and equities might be more positively correlated in the future. “It’s not our base case, but I think it is a downside risk scenario that it is fair to consider,” he says.

Gold: a momentum asset?

Beyond bonds, investors can use alternative assets to diversify their portfolio.

Infrastructure, property and even private equity can play this role to an extent, particularly for institutional investors. Part of the rationale is that private assets’ valuations are updated much less frequently than those that are traded daily on public markets, which reduces the overall volatility.

Investment trusts remain the best way for retail investors to access alternatives, but as listed vehicles they do not have this advantage: they can be quite volatile. Still, certain areas of infrastructure and property are especially focused on safe income streams rather than growth, and as such can act as diversifiers.

Another common diversifier, commodities, are not the domain of the average private investor – with the exception of gold.

Asset managers are divided on what role gold can play in a portfolio. In 2025, the gold price rose more than 50 per cent in sterling terms, far outperforming equities. The precious metal held up well as equity markets crashed in April as a result of Donald Trump’s so-called “liberation day” – and then it just kept going up.

Gold doesn’t have fundamentals and produces no cash flow. It’s all about sentiment. As Storm puts it: “It’s a momentum-driven asset. Supply is relatively stable, so demand is what really matters.”

In theory, when global investors are concerned about the global economy, they flock to bullion as a safe haven – so it should hold up well when equities struggle. But in reality things are a little more complicated than that; for example, if there’s a dash for liquidity during a crisis, the gold price is likely to fall with everything else, at least initially.

Additionally, investors are particularly nervous about this year’s returns. “Something that’s gone up 50 per cent in a year isn’t a safe haven, because if it’s gone up by that much it can fall by a significant margin as well,” says Morgan.

The price increase has been partly driven by gold purchases from central banks, which are less price sensitive than the average investor. Rees explains that together with a fractured geopolitical landscape, the sanctioning of Russian foreign exchange reserves following the invasion of Ukraine in 2022 played a major role. “That told you, or anyone in the world, that if you have assets in the wrong domicile, they can be taken from you,” he explains. So central banks are looking to diversify their reserves.

Storm argues that gold can still work as a hedge against certain specific scenarios – namely, an extreme equity shock or the debasement of fiat currencies (which can be caused, for example, by governments printing more money or running large deficits).

“But is gold a great long-term diversifier against inflation? Absolutely unproven. Is it a great diversifier against a minor growth shock? Absolutely unproven,” he says. And while the precious metal is traditionally almost uncorrelated to equities, that has not quite been the case in the past year.

“We’ve seen them sell off together and go up together. So is it as effective as a portfolio hedge, particularly when most of the risk in the portfolio is equity risk? Probably not right now,” Storm says.

Logically, the most straightforward ways to hedge against an equity downturn is cash and derivatives such as put options. But the latter are quite a complicated option for private investors, and should be handled carefully and by those with experience.

Your emergency cash fund should be treated separately from your diversification efforts, but if you want to reduce the equity risk in your portfolio and are not sure about long-dated bonds, you can also increase your allocation to cash or cash-like instruments (such as very short-term bonds), to reduce volatility and build a bigger buffer. For many, the best strategy will be to simply stay the course: keep in mind that if you have a long time horizon, it’s always better to stay invested and do nothing through the inevitable ups and downs.

More Investors’ Chronicle Big Reads

Investment Ideas of the Year 2026

How to invest in 2026

How to profit from self-driving cars

Four quality US shares to boost your portfolio

How to pick the right investment trusts

Aim 100 2025: The stocks to buy, hold and sell

Seven signs a share is a ‘ten-bagger’

Income Majors 2025: How the UK’s biggest dividend payers are faring

IC Top 50 Funds 2025: the best funds to buy

We’re fond of stock screens as a source of inspiration, but they are just the starting point for investment analysis. For instance, while our Aim quality shares screens isolate companies with solid operating margins and returns on capital, we must dig deeper to answer important questions about profit growth, cash generation and how much investors should pay up for the shares.

One Aim-traded stock that scores well on our quality screen is payment technology business Boku (BOKU), a company that had operational successes in 2025 and has a pathway to further profit growth if it executes its plans effectively.

A payments growth stock

Boku is a technology provider that helps merchants leverage the accelerating global trend of local payment methods, which in aggregate are overtaking globalised networks such as Visa and Mastercard in transaction volumes. Boku provides vendors with a single contract and technical integration to access hundreds of interfaces.

Throughout the world, and especially in fast-growing economies where people are less likely to have a traditional bank account, consumers are paying via direct carrier billing (whereby transactions made with mobile devices are added to phone bills), real-time account-to-account (A2A) transfers and digital wallets. Boku connects merchants not only to wallets from western providers including Google, Apple and Meta, but also other global giants Alipay, GrabPay and GoPay as well as real-time transfer systems such as India’s UPI, which allow transactions to be made from a phone using just a virtual payment address.

It’s not all about the big players. Boku’s custom connections with more than 300 payment methods raises the possibility of selling via more than 7bn consumer touch points in more than 90 countries.

Merchants such as Amazon, Microsoft, Meta, Google, Sony, Spotify, Netflix and Tencent use Boku’s platform to sell products and subscriptions, so it is well integrated. Supposedly, the competitive moat of Boku’s business model is complexity – overcoming technical and legal challenges across multiple system specifications and jurisdictions is too much of a distraction for a sprawling tech giant with its fingers in many pies.

That said, the market size opportunity (Boku estimates the value of mobile wallet transactions alone will be $9.6tn per year by 2028) may make Big Tech feel the effort of customising their own solutions is worthwhile, especially if advances in artificial intelligence (AI) and quantum computing simplify the task.

Analysts at Peel Hunt (one of Boku’s house brokers) argue that AI more generally is not an immediate threat to the company, and that its background effects could even boost growth. Boku’s proposition is solving regulatory friction with bespoke technical and settlement needs, notes Peel Hunt’s Damindu Jayaweera; and AI’s impact is more likely to be focused on accelerating the subscription economy, which could increase the total addressable market for Boku’s services.

Being cautious, however, competition and disruption are never risks that can be dismissed even when an advantage has been earned.

Leaving conjecture aside, there are reasons for optimism. FactSet consensus figures (based on eight analyst opinions) show Boku is expected to have delivered earnings per share growth of 14 per cent in the financial year ended December 2025, with 26 per cent growth to follow over the next 12 months. Double-digit operating margins for the past two financial years are testimony to progress and the respectable returns on invested capital (ROIC) – which, although not eye-catching are above the Aim median – allow Boku to meet our definition of a quality stock.

Full-year 2025 results are released in March, but the half-year numbers to June 2025 were positive, with underlying revenue up 27 per cent. The smaller digital wallet and A2A businesses are growing fastest, but direct carrier billing revenues were also up 15 per cent year on year in the first half. This is forecast to remain the largest segment by revenue out to the end of 2027, according to FactSet consensus data, which also implies sales should grow at a compound annual growth rate (CAGR) of over 20 per cent over that time.

Importantly, the amount of profit Boku is making per transaction has been rising, with its “take rate” up to 0.85 per cent in the first half compared with 0.81 per cent a year earlier.

Being a growth stock, with a potentially bright future, it is unsurprising that Boku commands a high price/earnings (PE) valuation. The reason for a next-12-months forward PE ratio of 28 times is because the market believes earnings will continue to grow rapidly beyond that timeframe.

Examining whether that rating is fair involves making sense checks with appropriate valuation models. As a growth company, it is unsurprising that Boku doesn’t pay a dividend, so dividend discount methods such as the Gordon growth model won’t work in this case.

Therefore, it would be more appropriate to value Boku based on either predicted free cash flow or the residual income (which means economic profit after taking into account investors’ opportunity cost) the shares could generate. Assessment of Boku’s cash position must take account of monies owed to merchants, although it does report its “own cash”, which was $87mn at the half-year 2025 stage – 9 per cent up on the 2024 full year.

Estimates of free cash flow per share are available for the years to 2027, with the annualised rate of growth being 13.7 per cent from the actual figure reported at the end of the 2024 financial year. It is notable that while cash flow per share is expected to come in lower for the 2025 year just ended, 2026 and 2027 are each predicted to see year-on-year growth at a rate of around 25 per cent.

Ideally, we would be looking at predicting free cash flow to the firm (FCFF), a measure derived from net income by adding back non-cash charges and after-tax interest expense, then subtracting fixed capital and working capital investments in the period. Per-share figures a few years out can then be discounted back at the firm’s weighted average cost of capital (WACC). We then subtract the value of debt per share to derive a fair value for the share price today.

Making fair estimates of these inputs is easier said than done, however, and there are a good many considerations even when preparing inputs for the calculation. Still, cash flow gives us a good way to consider a growth company from the perspective of a market price valuation, although it counts against Boku that it doesn’t look cheap versus similar UK-listed fintech companies.

For example, mobile-focused international payments business Wise (WISE), is rated on about 16 times the forecast of cash flow per share for its financial year ending in March 2028. By contrast, Boku is rated on 21.5 times its December 2027 forecast cash flows.

The premium rating Boku trades on shows a disconnect between its own growth narrative and what may yet prove conservative long-term growth forecasts. There are execution risks to the business model, but arguably these ought to be reflected in a higher required rate of return being used to discount the present value of future cash flows, rather than lowballing the estimates for cash flow growth rates.

Peel Hunt uses a high discount rate of 15 per cent as the required return in its models, which culminated in a fair value of £3.61 for the shares following publication of the last half-year accounts. The consensus figure on FactSet is £3.19, which still implies considerable upside to the £2.19 share price at the time of writing.

Investors can be forgiven for being circumspect, however. Many will remember getting burned on Bango (BGO), which made significant progress on its subscription bundling and payments packages but disappointed against revenue targets. It also repeatedly got caught out by higher than expected operating costs and research and development expenditure.

Unexpected costs are the type of surprise that could derail a business like Boku, but it has made strategic progress and has a significant opportunity to grow its revenues just from existing merchants’ use of its technology. Recent highlights include the agreement with Australian graphic design software company Canva, which it signed last August, and the addition of 60 new connections in total for merchants in the first half of 2025.

Along with other UK-listed tech companies, Boku’s share price dropped in the last quarter of 2025, but this runs counter to analysts upgrading views on its longer-term prospects, with consensus for the 2028 financial year earnings seeing a particularly strong jump.

Long-run growth rates are often related to expected economic growth in countries where revenues are earned. Although Boku makes 57 per cent of its revenue from the Asia Pacific region and 39 per cent from Europe, Middle East and Africa, the sustainable growth rate being assumed beyond the next five years for earnings and cash flows is likely to be conservative. That may be prudent given the heightened geopolitical uncertainty and the fact demographics are emerging as a challenge even in many developing countries.

Still, technological adoption in these countries is driven by mobile, and there will be a growing addressable market of consumers. They’ll have disposable income and demand flexibility to make purchases and international transactions without needing to keep details such as bank cards saved online. Boku meets these needs and, importantly, it has also been successful in facilitating a user experience that doesn’t compromise Big Tech vendors’ insistence on smooth single-click, in-app purchases. These make up around 40 per cent of Boku’s payment volumes, with traditional checkout experiences accounting for just 5 per cent.

Further innovations Boku has invested in include a cross-border settlement platform offering merchants a foreign exchange liquidity network and a banking footprint. Future areas of focus could include the ability for merchants to make payouts to local influencers – an offering that fits modern marketing techniques and which is an intriguing prospect.

The bull case for Boku shares rests on confidence that the extraordinary growth phase used in valuation models can be extended. Should there be more signs of good news ahead of the company’s results, Boku could benefit from earnings upgrade momentum.

In the meantime, the company has been putting some of its excess ‘own cash’ to use on buybacks. On 2 January, Boku announced that the board had approved the repurchase of up to 5 per cent of outstanding common stock. Purchases will be added to the 2.1 per cent of shares already held in treasury, with the plan being to meet obligations to warrant holders or staff remuneration programmes without diluting ordinary shareholders in future. Clearly, the board thinks the shares are cheap given Boku’s opportunity. They may have a point.