Picking the best ETF. Working out whether US stocks are too expensive. Deciding whether rates will fall. Selling before the next recession and buying back at the right moment. Investing can turn into a long sequence of decisions. Lazy portfolios start from an almost opposite idea: take some of the most important decisions in advance and try, as far as possible, not to have to take them again when markets put the investor to the test.
It doesn’t mean building a portfolio and forgetting about it. And above all, lazy doesn’t mean careless. It means trying to build an allocation diversified and understandable enough to be maintained over time with few rules, few instruments and limited intervention. How do you build a portfolio without knowing what will happen tomorrow?
Why are they called «lazy»?
«Lazy portfolio» can be a misleading name if it makes you think of a portfolio built with little care. It’s almost the opposite.
The name is mostly linked to the American financial journalist Paul B. Farrell, who helped popularise the expression «lazy portfolios» to describe simple, diversified, low-maintenance portfolios. The Bogleheads community itself credits Farrell with popularising the term.
The «laziness», then, comes after the portfolio is built. You first decide how to distribute the capital, which rules to follow and how to rebalance. Then you try to intervene as little as possible. Lazy doesn’t mean thinking little about the portfolio. It means building it so you can think little about it afterwards.
Weights come first, instruments second
When people talk about investing it’s easy to focus on the product: which ETF, which index, which TER, which issuer. Those are important questions. But they come after a more fundamental decision: how do you distribute your wealth across assets that behave differently?
A portfolio that is 90% stocks remains essentially an equity portfolio even with the cheapest ETFs on the market. A portfolio combining stocks, bonds and gold asks a different question instead: what role should each component play, and how much capital do we want to assign it? This is the asset allocation problem. Vanguard, in its investor education material, shows how much a portfolio’s historical behaviour can change simply by shifting the stock/bond mix, using series that reach back to 1928.
The middle value of each column is the average annual return: the mean of individual calendar years, not the compound annualised return. And the extreme columns tell two things that are easy to miss. The best year of the 100% bond portfolio (+32.6%) was higher than the best year of the 50/50 (+31.1%). And even the 100% bond portfolio closed one year at −12.9%. The labels «conservative» and «aggressive» describe a long-term profile, not the behaviour of every single year. And bonds do not mean the absence of losses.
What a lazy portfolio is
There is no formal definition of a lazy portfolio, and no precise number of instruments that turns a portfolio into a «lazy» one. The term is normally used to describe portfolios that are relatively simple, diversified, built on pre-set asset allocations, often implemented with low-cost index funds, held for long periods and rebalanced by fairly simple rules.
«Lazy», then, mostly describes the maintenance required, not the rigour with which the portfolio was designed. What they share is not necessarily what they hold. It’s what they try to avoid: the need to continually reinvent the portfolio.
A lazy portfolio can even be 100% equity
«Lazy» is not a synonym for conservative, balanced or multi-asset. Even a 100% equity portfolio can be perfectly lazy: it can follow a simple rule, be broadly diversified within equities and be held for decades without trying to predict the market.
It would, however, be a portfolio with a very different risk profile from a 60/40, a Permanent Portfolio or a Golden Butterfly. So the question is not how conservative a portfolio must be to earn the «lazy» label, but whether its asset allocation is consistent with the investor’s goals, time horizon, future capital needs and capacity to bear losses.
That capacity has at least two dimensions. One is economic: how much a loss can compromise a goal or force selling at the wrong time. The other is psychological: how much the investor can genuinely tolerate swings and losses without abandoning the plan. And here lies a limit that questionnaires struggle to measure: someone who has never been through a very deep drawdown with real money can imagine how they would react to −30%, −40% or −50%, but can hardly know for sure before living it.
Why diversify, if it might mean earning less?
Diversifying doesn’t mean trying to maximise returns in every period. Adding assets that behave differently can mean giving up part of the return that, in hindsight, the best asset class would have produced. The reason is different: reducing how much the overall outcome depends on a single source of return, a single region, a single factor or a single economic scenario. Diversification doesn’t guarantee a profit and doesn’t prevent losses; it changes how risks are distributed within the portfolio.
| What you seek | Why it can matter |
|---|---|
| Reduce concentration | Avoid the outcome depending too heavily on one asset class, region or strategy. |
| Limit the depth of losses | A smaller loss needs a smaller percentage gain to get back to the initial capital. |
| Reduce time underwater | A shallower drawdown can — but need not — translate into a shorter recovery. |
| Make the plan bearable | A portfolio that works on paper must be maintainable through the worst phases too. |
| Accept a possible opportunity cost | Diversifying assets can underperform the best asset for years or decades. |
Diversifying doesn’t mean seeking the portfolio that will lose least in every crisis. It means accepting that something may do worse, so that the overall result depends less on one single thing going right.
The mathematics of losses
There is also a purely mathematical reason why the depth of a loss matters. Losses and recoveries are not symmetric: after a loss, the return needed to get back to the starting point grows faster and faster. After a −50%, a subsequent +50% is not enough: €10,000 becomes €5,000 and then €7,500. Getting back to €10,000 takes +100%. This is not a backtest and contains no forecasts: it’s simply mathematics.
And recovering nominally is not recovering in real terms
Even returning to the initial capital tells only part of the story. If €10,000 falls to €5,000 and later returns to €10,000, the nominal recovery is complete. But if years have passed and prices have risen, those €10,000 no longer buy the same things. For example, with average inflation of 2% for five years, preserving the initial purchasing power of €10,000 would require about €11,041.
This is one of the reasons why, throughout the series, we will distinguish where possible between nominal and real returns, and pay attention not only to the max drawdown but also to the time needed to recover it.
The hardest decision may be the one we won’t take
Imagine a portfolio that decides in advance to hold 60% stocks and 40% bonds. After a strong equity rally it might become 70/30. Rebalancing doesn’t require working out whether stocks will keep rising. The rule was decided beforehand.
The same principle becomes even more interesting during a crisis. When an asset loses 30%, the market hands us new information and enormous psychological pressure at the same time. The lazy portfolio tries to reduce the number of these discretionary decisions. It doesn’t eliminate uncertainty. It tries to decide in advance how to behave when uncertainty arrives.
A family, not a recipe
There is no such thing as «the» lazy portfolio. Over time many allocations became popular, born from very different philosophies. The Three-Fund Portfolio associated with the Bogleheads tradition seeks extremely broad diversification through a few equity and bond funds. Harry Browne’s Permanent Portfolio splits wealth across four components designed for different economic conditions. The Golden Butterfly takes part of that philosophy and modifies it to increase exposure to economic growth. The portfolio attributed to David Swensen introduces further sources of diversification. The Coffeehouse Portfolio splits the equity sleeve into several segments.
We won’t rank them straight away. First we’ll try to understand why they exist.
There’s a catch: most lazy portfolios speak American
This is where the series becomes interesting for a European investor.
Take a recipe containing US Total Stock Market, US Small Cap Value, Long-Term Treasuries, Short-Term Treasuries and Gold. For a US investor the meaning is fairly natural. For an investor who lives, saves and will mostly spend in euros, much less so.
Must a US Treasury remain a US Treasury? Or is what matters the function that bond performs in the portfolio? Should the US domestic equity sleeve become European equity, or global? Is dollar exposure a desired component of the portfolio, or simply a consequence of its American origin? Swapping one ticker for another is not enough.
We won’t translate tickers. We’ll translate functions.
This will be the rule of the series. For each lazy portfolio we will start from the original version. We will try to reconstruct its origin, its asset allocation and above all the economic role of each component. Only then will we try to build a possible European implementation using UCITS ETFs or, where necessary, other listed instruments available to the European investor.
A Rebalix adaptation will not be presented as the original author’s portfolio. It will be an experiment: what happens if we try to preserve its logic using instruments available to a European investor?
Before looking at results, we will freeze the rules
There is a particularly insidious risk when building a portfolio by looking at the past: try a combination, look at the return, change an ETF, try again, change a weight and try yet again. To reduce this problem we will follow the opposite order: reconstruct the original version; identify the components’ functions; define the European adaptation; choose the instruments; freeze the allocation; only then compute the results.
We won’t change an ETF because we dislike the backtest. We won’t swap an index because another one would have produced a higher CAGR. If the result is mediocre, mediocre it stays.
The rear-view mirror
A backtest is an extraordinarily useful tool and an extraordinarily easy one to misread. A backtest looks in the rear-view mirror. And for a European investor that mirror has at least two big problems.
First problem: the European mirror is short
Vanguard can show asset allocation analyses starting in 1928. Almost a century. Within that period we find radically different market conditions: the Great Depression, the inflation shocks of the Seventies, the Volcker rate hikes, the 1987 crash, the 2000–2002 tech bubble, the 2007–2009 global financial crisis, the 2020 pandemic and the 2022 inflation-and-bond shock.
But Vanguard doesn’t own an ETF born in 1928. To reach that far back it builds its series by chaining different indexes over time. That’s a legitimate methodology when disclosed, but it answers a different question. Asset class backtest: how could this asset allocation have behaved historically? Instrument track record: how did these specific, actually investable instruments behave? The first lets you look much further back. The second needs fewer reconstructions, but often has a far shorter history. Throughout this series we will always keep the two apart.
Vanguard’s methodology notes add a precious detail: in that series, stocks are 100% US until 1969 (then 60% US / 40% international) and bonds are 100% US until 1990 (then 70% US / 30% international). Even when we say «almost a century of data», we must ask what that data really represents: for most of the sample, the viewpoint of an American investor measuring in dollars. We’ll come back to this — it’s the same problem as lazy portfolios that «speak American».
1929 may simply not exist in your numbers
Suppose a European portfolio shows a max drawdown of −14.8%. It’s very easy to read it as: this portfolio loses at most about 15%. But that’s not at all what the figure means. It means: in the historical period observed, the worst drawdown was −14.8%. If our history starts in 2015, that portfolio never went through 1929, the Seventies, 1987, the dot-com bubble or 2008. We don’t know what would have happened.
A max drawdown without the period it comes from tells only half the story.
Second problem: even a long mirror looks the wrong way
Imagine having a hundred years of perfect data. We know how the portfolio behaved through depressions, wars, inflation, deflation, bubbles, recessions and financial crises. One problem remains, and it cannot be eliminated: all those crises have already happened. The next one hasn’t. The future is under no obligation to reproduce the same relationships between stocks, bonds, inflation, interest rates and currencies observed in the past.
Will the dollar that protected us yesterday protect us tomorrow?
In several phases of severe financial stress the US dollar strengthened against other currencies. For a eurozone investor holding unhedged US assets, the currency move can soften part of the loss when the result is measured in euros. But it is not a law of finance. And above all, we don’t know whether it will happen in the next crisis.
If we replace US Treasuries with euro-denominated government bonds, we can reduce the European investor’s currency risk. But we might also remove a source of diversification that helped in some past crises. Or we might remove a risk that would have hurt in the next one. We cannot know in advance.
Making European a portfolio that survived many crises in the United States does not automatically preserve its future behaviour. But keeping the American version unchanged does not automatically preserve it better either.
So what is a backtest for?
If it can’t tell us which portfolio will win, why do it? Because it can answer many other questions: how volatile it was; what the maximum observed loss was; how long it took to recover it; how the components behaved together; how often it closed a year at a loss; how it reacted to the regimes in the sample; how much it cost to maintain. As long as we don’t turn a description of the past into a forecast.
Backtests help us understand how portfolios worked. They don’t tell us how they will work.
How we will analyse lazy portfolios
To make the coming instalments comparable we will use a common methodology. For each portfolio we will first show the original version, clearly separated from any Rebalix adaptation. For the European version we will declare the instruments used, the index tracked, the stated costs and the actually available period. Performance will be computed with a methodology consistent across portfolios, and rebalancing will follow a rule defined before the analysis.
Among the metrics: annualised return, volatility, maximum observed drawdown, recovery time, positive and negative years, best and worst year, Sharpe and Sortino, rolling returns, growth of an initial capital and weighted costs. And above all, we will always show since when we are measuring.
We won’t look for the winner
At the end of the series we will put the portfolios side by side. But we won’t build a ranking like «The best lazy portfolio of 2027». One portfolio may have had the highest CAGR simply because the observed period favoured the assets it held. Another may have shown a smaller drawdown because our history doesn’t contain the regime that would have hurt it most.
We will try instead to understand which trade-offs each allocation buys: return, risk, recovery time, simplicity, diversification, costs and the concrete possibility of sticking with it when it stops working as hoped.
Next instalment: the Golden Butterfly
We will start with a very particular portfolio: five components, five identical weights. 20% US large cap blend equity, 20% US small cap value equity, 20% long-term US Treasuries, 20% short-term US Treasuries, 20% gold. It’s the Golden Butterfly.
But we invest from Europe. What does that 20% of long-term Treasuries mean? Should we keep the dollar? What can represent small cap value? And can we build it all with instruments actually available to the European investor without turning the Golden Butterfly into something that no longer has anything to do with the original?
In the next instalment we will do exactly that. First we will build the portfolio. Then, and only then, we will look at what the numbers say.