I listened to a podcast where a financial advisor walked through his Goal Based Investing philosophy with a real case. The details were refreshingly specific: a couple, 300k saved, one kid headed to university in 15 years, and a retirement target of 5,500/month at age 60.
His recommendation? A 90/10 portfolio. 90% stocks, 10% bonds.
I’ll set aside my confusion about Goal Based Investing, I thought the whole point was a separate portfolio for each goal. Maybe I misunderstood. Maybe they were keeping it simple for the podcast. Moving on.
What this case did was get me thinking about the university funding goal specifically. I’d argue it’s one of the hardest goals to execute with a traditional stock/bond portfolio.
Here’s the trap: 15 years feels like forever. Long enough that you convince yourself you can ride out volatility with a heavy equity allocation. And you’re not wrong…until you’re three years from the deadline and the market decides to remind you that it doesn’t care about your kid’s enrollment date.
Two problems deserve attention here.
First: sequence of returns risk. This isn’t just a retirement problem. It hits any goal with a fixed deadline. A bad run of returns in years 13-15 can wreck a plan that looked great for the first decade.
Second: the illusion of de-risking with bonds. The idea that gradually shifting into bonds “protects” you sounds intuitive. The data is less reassuring, especially if you consider an inflation shock.
For this analysis, I used PortfolioCharts. They don’t have a Monte Carlo simulator, but I’d rather have honest historical data than a flashy simulation built on questionable assumptions.
And speaking of Monte Carlo, let me be blunt. These simulations are only as good as the model underneath them. If you’re working across multiple asset classes, you need to preserve correlations and ideally the mean-reverting behavior of (some) returns. And fat tails. The dirty secret is this: either you use historical data, which puts you right back at PortfolioCharts, or you build a proper full model, the kind that no retail investor will ever have access to. More data points from a bad model aren’t better than fewer data points from reality.
I will compare the results of an 80% stocks, 10% bonds, 10% REITs portfolio (the advisor mentioned he likes REITs), “SBR” from now on, against PortfolioCharts’ own champion, the Golden Butterfly, “GB”. For both portfolios, the home country is set to Italy.
Sequence of Return Risk
The couple has a target of €125k for their son’s university budget. I ran a simple test: how long would it take to reach that target saving €2k a year.
GB:
SBR:
Look at the distribution of outcomes between the two portfolios. Remember: the only thing that matters is hitting the number (it is goal-based, innit?).
At year 27, the GB has roughly 4-5 cases that fall short. The SBR? One sequence gets you there in 17 years, two in 18, but the spread of outcomes is dramatically wider. The cloud of lines for the SBR only starts to converge around year 21.
Now for the elephant in the room.
Each line is a historical sequence. Many of them overlap. They are not independent. This is the fundamental limitation of PortfolioCharts, and you should not pretend otherwise.
That said, from 1970 to today, markets lived through a lot: high inflation and low inflation, booms and stagnation, oil shocks and tech bubbles. I wouldn’t bet my retirement on this dataset alone, but the intuition it provides is still valuable. And in many cases, it’s the best we have.
Because here’s the thing about “improving” on historical data with Monte Carlo: it only works if you really know what you’re doing. You can’t just plug in 4% expected real return and 17% vol for stocks and call it a distribution. The quality of the output depends entirely on how the simulation is built.
What I love about PortfolioCharts is that it uses real historical inflation combined with real asset class returns, period by period. Almost every simulation tool I’ve used lets you customize inflation…but then applies that single rate uniformly across the entire analysis. That’s not modeling. That’s a flat assumption dressed up as rigor. The world doesn’t work in straight lines.
Plugging in 2% real return for stocks and 3% inflation might look conservative on paper. But a model that treats inflation as a constant is useless for understanding what actually happens to a portfolio. Inflation has regimes. Different asset classes respond differently to each one. Some hedge well. Some don’t. PortfolioCharts might be the only retail tool that lets you stress-test against a real high-inflation regime.
One more thing worth noticing: the higher the annual contribution, the better the GB looks relative to the SBR. At €3k/year, the GB range tightens to years 18-25. The stock-heavy portfolio? 14-27. Make of that what you will.
Now look at the outcome distribution at year 31. The SBR has five positive outliers: the fat tail that pulls the mean higher for a stock-heavy portfolio and makes it look attractive on average. That upside is real. But it isn’t free.
To get there, you have to hold through volatility without panicking. You have to not need the money before the market decides to cooperate. The cost isn’t visible in the return distribution, but it’s there. Volatility is the price of admission. Most investors say they’ll pay it. Fewer actually can do it.
The Ulcer Index is the best way to visualise the relationship between portfolio volatility and sequence of return risk:
The GB lost more than 10% in a single year only in 3 cases.
The SBR lost more than 20% in a single year in 5 cases and more than 10% in 6.
After 3 years, it was below 10% in 9 cases vs 1 for the GB.
It is hard to say “I reached my target” when one bad year is all it takes to undo years of progress. Sure, the 125k ‘university target’ is spread over 5 years, but a 12.5k annual hole is no easy thing to plug.
Another way to see it is with the Target Accuracy chart. First, the GB:
and the SBR:
Another way to see it: to reach the same average, the GB investor has to invest 11.8k instead of 10k. The “low vol” portfolio comes with a non-negligible price attached…in exchange for a more reliable baseline.
Even after 15 years, the baseline of the SBR portfolio is still below the min of the GB.
De-risking
The classic fix for everything I’ve described is the glide path. As you close in on your goal, you de-risk, rotating into a heavier bond allocation once you’ve built enough buffer to lock in the target. This is exactly what large Defined Benefit pension funds do, and I’m not dismissing it. The establishment is happy with this solution, so why your financial advisor shouldn’t?
But here’s the uncomfortable truth: it doesn’t work in a world where inflation can spike any year. Unless you go to the financial plan sponsor, in this case you – the client, and ask for more money 😉
As I’ve written before, there is no single-asset solution to de-risking in a real sense (pun intended). You need a diversified portfolio to hedge against sequence of return risk that is inflation-adjusted, not just nominal.
The reason this keeps getting ignored is simple: inflation spikes have been rare enough in recent decades that a financial advisor can frame the exceptions as black swans. No one saw it coming. Don’t fire me. And then, having said that, he goes right back to recommending bonds as the de-risking tool…fully aware that the last crisis exposed its limits, fully committed to the same playbook.
The alternative is genuinely harder to sell. It requires more client education. It produces more tracking error in normal years. And yes, correlations that look stable can break exactly when you need them most.
But I struggle to understand how it is responsible to bury your head in the sand and pretend the risk doesn’t exist. Convenience is not a risk management strategy.
Too pessimistic – Save too much
Here’s the part I find hardest to follow.
The advisor admits that past equity returns may have been exceptionally high. Fair enough, that’s a reasonable concern. So he adjusts: lower expected returns in the Monte Carlo, and he targets a low percentile of the distribution rather than the mean. The mean becomes the optimistic case. The 25th-ish percentile becomes the plan.
But then why are you running a high-volatility portfolio in the first place?
The entire justification for loading up on stocks is the return premium. You accept the volatility because of the upside. But if your financial plan deliberately refuses to take credit for that upside, if you’re explicitly planning around the tail, then you’ve kept all the pain and given away the gain. You’ve chosen the worst of both worlds.
If you’re targeting the 90th percentile success rate in a Monte Carlo, you are, by construction, planning for the 10th percentile outcome. If that’s genuinely your target, then optimize the portfolio for that target. A lower-volatility, better-diversified portfolio will get you to the 10th percentile more reliably, with less suffering along the way.
There’s another issue hiding in plain sight. When a financial plan tells you “you’re fine,” it usually means “you won’t run out of money in bad scenarios.” What it doesn’t say is “you’re NOT optimizing your lifetime consumption”: which is, arguably, the actual goal.
The irony is brutal. You end up saving more than you need to, but you can’t touch the capital because it’s mentally earmarked for the future. This is especially acute for the retirement goal. If something goes wrong today — a health scare, a job loss, an opportunity — and you decide to raid those funds, you will spend years feeling guilty about a retirement shortfall that, statistically, was never going to materialize.
Your advisor has constrained your present so you arrive at retirement with too much…and a portfolio volatile enough to keep you anxious the whole way there. But don’t worry: he’s available to talk you off the ledge whenever volatility spikes. That’s what the 1% annual fee is for. You’re paying for the emotional management of a problem he created?
And then you reach retirement…and the same conservative logic kicks in again. Plan for the 10th percentile. Spend cautiously. Die with a pile of unspent capital.
I genuinely don’t understand who this is optimized for.
Portfolio Visualizer
The advisor mentioned his practice uses Portfolio Visualizer as a planning tool. It’s been a while since I’ve used it: their decision to put 95% of the platform behind a paywall was what kickstarted Testfol.io in the first place (which, unfortunately, seems to be heading down the same road these days).
I went back, built a Monte Carlo simulation with the numbers he shared during the interview. I can only model a hypothetical US-based client rather than an Italian one, but look at what comes out.
The financial goals:
The SBR portfolio:
https://www.portfoliovisualizer.com/financial-goals?s=y&sl=7g7SjB6GzmM8y0TFKpx0rQ
The GB:
https://www.portfoliovisualizer.com/financial-goals?s=y&sl=EvsNsF4FmIxZLhdw2XZfD
I used plain vanilla assumptions, you can check them through the link. I’d be genuinely curious to see what set of assumptions would lead anyone to conclude that a stock-heavy portfolio outperforms a diversified one here.
Would the real macro trader please stand up?
In every simulation I ran, I used a plain vanilla, market-cap-weighted allocation for the equity sleeve. During the interview, the advisor mentioned he typically adjusts away from market-cap weights for the usual reasons: home country bias to reduce FX risk, trimming exposure to expensive markets, layering in small and mid caps, emerging markets, and so on. I may not be remembering every detail of that specific conversation accurately, but the broader point stands: this kind of tinkering is standard practice across the independent financial advisory world.
You know who else engages in these tilts? Macro traders. You know who makes money doing it? Almost no one.
Think about what you’re implying when you hire someone who actively manages factor exposures and country tilts. You need them to be smart enough to understand market dynamics…but not so smart that they’d rather be running money at a hedge fund. What are the odds you find that person? And what are the odds they stay in retail advisory if they actually have the skill?
As I’ve written before, tilting toward European stocks as an FX hedge is a poor trade. Yes, it reduces currency exposure, but it drags in a whole set of unintended consequences: different sector weights, different factor exposures, idiosyncratic risks. A simple FX overlay would accomplish the same thing without the collateral damage. But I suppose you need to pay someone 1% a year to not know that. 😛
What I am reading now:
Follow me on Bluesky @nprotasoni.bsky.social
0 Comments