/Wealth

**Asset Allocation Is What Actually Builds Wealth (Not the Stock You Think Is Genius)** Asset allocation is one of those things everyone talks about in theory and almost nobody applies seriously in practice, and I say this after years watching friends and acquaintances dump money into things because they "heard about it" or because someone on Reddit swore it was the next Tesla. Let me be blunt from the start: picking the right stock is, for the overwhelming majority of people, an ego exercise disguised as financial strategy. And asset allocation, that boring, unglamorous business of deciding what percentage goes into stocks, what goes into bonds, what stays in cash, is what actually separates people who build wealth from people who spend twenty years shooting in the dark. This isn't just my opinion. There's an old study, from 1986, by Brinson, Hood and Beebower, which became known (sometimes misquoted, it must be said) for suggesting that over 90% of the variability in a portfolio's returns comes from asset allocation rather than individual security selection or market timing. There's plenty of academic debate over whether that number is being interpreted correctly — and it partly isn't, because explaining the variance of returns over time within a portfolio is one thing, saying 90% of absolute return comes from there is another. But even the fiercest critics of that paper don't disagree with the core idea: how the portfolio is structured matters far more than the individual bets you put in it. This bothers people. It bothers them because nobody wants to hear that all that stock picking, all those hours reading Apple's quarterly reports or trying to figure out whether Nvidia is expensive or cheap, is worth relatively little compared to the simple decision of "I'll hold 70% stocks, 30% bonds." It's demoralizing. It's like telling someone who loves cooking that the final flavor of the dish depends more on the quality of the base ingredients than on plating technique. Technically true, emotionally unbearable. Let me explain why this happens, mechanically, because I think without understanding the mechanism people keep ignoring the advice. When you buy an individual stock, you're exposed to brutal idiosyncratic risk. Enron collapsed. Wirecard, more recently, evaporated in two days after it turned out 1.9 billion euros on the balance sheet simply didn't exist. This isn't market risk, it's company-specific risk, and in portfolio theory it's diversifiable — meaning you shouldn't be compensated for taking it on, because you can eliminate almost all of it just through diversification. Except most people don't diversify enough. I have a friend, a software engineer, brilliant at what he does, who back in 2019 put maybe 60% of his savings into three tech stocks because he "knew the sector." One of them did well. The other two are, even today, still recovering from what they lost in 2022. His portfolio's aggregate return over those three years fell well short of a plain global ETF, and he knew more about tech than 95% of people managing funds. Sector knowledge doesn't compensate for lack of structure. And here's the point almost nobody explains properly: asset allocation isn't about picking the "best" assets. It's about managing the correlation between them. When stocks fall, quality bonds tend, historically, to behave differently — not always inversely correlated, that's a simplified myth floating around, but with a correlation low or unstable enough to cushion the shock. In 2008 this worked beautifully. In 2022 it failed spectacularly, because both stocks and bonds dropped at the same time due to the Fed's aggressive rate hikes. That caught a lot of people off guard, including 60/40 portfolios that prided themselves on being "protected." So no, asset allocation isn't a magic formula that always works the same way. It's a system of probabilities, not certainties, and anyone selling it to you as a certainty is either lying or doesn't understand the subject. Which brings me to something that irritates me particularly: the sheer amount of content online, especially in Portuguese, that treats asset allocation like a fixed recipe. "60% stocks, 40% bonds, done." Nothing is done there. The right allocation depends on your time horizon, your risk tolerance (which is psychological, not mathematical, and this is crucial), your risk capacity (which is objective and financial), and factors that shift throughout life. A now-retired wealth manager I knew used to tell me the biggest source of value destruction wasn't the wrong allocation on paper, it was the right allocation on paper that the person then couldn't psychologically stick with during a 34% market drop. They'd sell at the bottom. Always. He'd seen it hundreds of times over his career. Which brings me to a distinction I think is fundamental and rarely well explained: risk capacity versus risk tolerance. Capacity is objective — how much time you have until you need the money, what other income sources you have, whether you have expensive debt to pay off first. Tolerance is subjective, how much you can watch your portfolio drop 20% without panicking and selling everything at three in the morning because you can't sleep. A 28-year-old with a stable job has extremely high risk capacity, objectively — could sit at 90% stocks or more without much technical problem. But if that person's temperament is to panic at every 8% correction, their tolerance is very low, and in that case a more conservative allocation than the mathematically "optimal" one might actually be better, just to make sure they don't do something stupid at the wrong moment. This is something the pretty Markowitz optimization models simply don't capture, because they assume a rational investor who doesn't exist outside the textbooks. Speaking of Markowitz — modern portfolio theory, from the early '50s, with its efficient frontier and all that, is gorgeous on the whiteboard. In practice it's a nightmare to implement because it depends on estimates of future expected returns, volatilities and correlations that nobody can predict with precision. The inputs are the whole model's Achilles' heel. Small changes in estimates generate completely different allocations, a phenomenon known as "error maximization" — the model isn't robust, it's hypersensitive to estimation error, and it amplifies that error instead of smoothing it out. I spent an entire afternoon once, a few years back, playing with a spreadsheet trying to replicate this, just varying expected-return inputs by half a percentage point to see how much the optimal allocation shifted. It shifted dramatically. That scared me more than anything else I've read on the topic, because I realized how much of the "scientific rigor" around asset allocation rests on quicksand. Not that this invalidates the core idea. It remains true that diversifying across asset classes with different behaviors reduces risk without proportionally sacrificing expected return. It's just that the surgical precision a lot of people think exists, that talk of "the optimal allocation is 63.4% global equities, 24.1% bonds and the rest in REITs," is theater. Nobody knows that with that precision. What we know, with reasonable historical confidence, is the general direction: more stocks, higher expected return and more volatility long-term; more bonds, more stability and lower expected return; some international exposure reduces the risk of being too dependent on a single country's economy (something people in Portugal tend to forget, with a massive bias toward Portuguese stocks and local real estate, so-called home bias, a classic and very common mistake, especially among older investors who grew up in a more closed financial culture). Let me digress for a moment, because something just came to mind. A while back I was reviewing a relative's portfolio, someone in their early 60s, and found it was almost entirely in fixed-term deposits with a little bit of BCP stock bought back around 2006, still sitting on a massive loss from that. The issue wasn't just the lack of diversification, it was the total absence of any allocation logic whatsoever. There was no decision there, just inertia accumulated over decades. And this is more common than people think — people don't have bad asset allocation, they have no allocation at all, which is actually worse, because at least a bad conscious decision can be fixed with a conversation. Inertia is harder to dismantle because it doesn't look like a decision, it looks like "just how things are." Back to the main thread: another point I think is brutally underrated is the cost of rebalancing, both the direct financial cost and the psychological cost. Rebalancing a portfolio, meaning selling what went up and buying what went down to get back to your target weights, is counterintuitive to the bone. It goes against the instinct to chase what's performing well. And it's precisely because it goes against that instinct that it works, most of the time — it's a systematic way of selling high and buying low without having to guess anything. But almost nobody does it in a disciplined way. Either they rebalance too late, or never, or they do it so frequently that transaction costs and tax implications eat the gains. In Portugal, with capital gains tax on stocks and ETFs running around 28% (with nuances depending on the instrument type and holding period, details worth paying attention to), every unnecessary rebalance carries a real cost most people ignore until they actually do the math at year-end. I also think it's worth talking about something rarely said out loud: the "ideal" asset allocation changes over time, but not in the mechanical way lifecycle or target-date funds suggest. That old "110 minus your age in stocks" rule is a crude heuristic, useful as a starting point, but dangerously oversimplified if applied without thinking. Someone at 55 with a solid guaranteed pension and no dependents is in a completely different situation from someone the same age who needs that money to live on in five years with no other safety net. Age is a poor proxy for what actually matters, which is the specific time horizon of each financial goal. And people usually have several goals with different horizons at once — retirement thirty years out, a house down payment in four years, an emergency fund for tomorrow — and try to cram all of that into a single allocation, which makes no sense at all. Ideally there should be different allocations depending on the goal and timeframe, a kind of mental accounting done well instead of done badly. This relates to something I read years ago, in a William Bernstein piece if I remember right, about the idea that asset allocation exists mainly to manage the risk of being forced to sell assets at the worst possible moment. It's not about maximizing return in the abstract. It's about making sure that when you need the money, you're not forced to sell stocks in a market down 40% just because you lost your job or had a medical emergency. That's why the cash and short-term bond component is so underrated — it's not there to generate spectacular returns, it's there to give you optionality and peace of mind during the bad stretches, which indirectly protects long-term returns by preventing panic selling. And maybe this is where I want to get a bit more cynical, because I think the financial industry, especially the more traditional wealth management side, has a structural incentive to make this more complicated than it needs to be. The more complex asset allocation appears, the more justifiable it becomes to charge management fees of 1.5% or 2% a year to "optimize" something that, for the vast majority of individual investors, is solved reasonably well with three or four well-chosen low-cost ETFs rebalanced once a year. I'm not saying professional management never has value — it does, especially in complex wealth situations, specific tax matters, estate planning, that kind of thing. But for the average investor with a twenty- or thirty-year horizon and relatively simple goals, the complexity being sold often serves the seller more than the buyer. That's a strong opinion, I know it is, and there are excellent advisors out there who'd disagree vehemently, and rightly so in some specific cases. One technical point I particularly like, and which I almost never see discussed outside nerdier circles: asset allocation interacts oddly with inflation, and that interaction has shifted lately. During the low-inflation decade, roughly 2009 to 2021, bonds did the diversification job pretty well because the main macro risk was deflationary or weak growth, and in those scenarios bonds rise while stocks fall. But in a high, unexpected inflation environment, like we saw in 2022, both asset classes suffer at the same time, because rate hikes meant to fight inflation hit both existing bond prices and stock valuations, especially the more speculative, growth-oriented ones. This means the correlation between stocks and bonds isn't a constant of nature, it's regime-dependent on the macroeconomic environment we're in, and a well-thought-out allocation should account for that, perhaps with some exposure to real assets like direct real estate (not REITs, which behave more like stocks), commodities, or inflation-linked bonds. I'm not saying people should rush out and panic-buy gold, far from it, but ignoring inflation risk entirely because "the last fifteen years weren't a problem" is the kind of naive extrapolation error financial history punishes regularly. I think it's worth ending on something more practical and less theoretical, because I know staying purely in analysis without landing anywhere is frustrating. In practice, for most people who ask me about this — and quite a few have, over the years, friends, family, people who reach out because of the sites I run — the advice I give isn't sophisticated. Define an honest time horizon for each goal, pick a stock/bond ratio you can actually stick with emotionally even during an ugly drop (and this requires some uncomfortable introspection, because most people overestimate their own risk tolerance until they see the account statement drop for real the first time), diversify geographically so you're not held hostage by a single market, and then basically don't touch it much, except to rebalance once a year or when the weights drift meaningfully from target. Boring. Not sexy at all. But it's what separates people who reach 60 with a solid nest egg from people who spent thirty years jumping from one financial fad to the next, always arriving late to every trend. It's no accident that Warren Buffett, despite all the fame tied to individual stock picking, left instructions in his will for his wife's fortune to be invested mostly in low-cost S&P 500 index funds, with a small portion in short-term bonds. He, who spent his whole life stock picking with extraordinary results, knew perfectly well that for most people, including his own family, simple, disciplined structure beats individual sophistication almost every time. That says a lot. More than any thirty-page academic report full of regressions.