Taking Advice from Algorithms: Why the Messy Line Matters

You know what real life looks like? Messy. But you wouldn't know it from most financial plans – not algorithmic ones, anyway.

Most advice out there comes as a straight line. A tidy formula. Clean inputs, clean outputs. Do X, get Y! Save this percentage, retire at that age. Follow these steps, achieve this outcome.

But real life is messier. It's a chaotic tangle of loops and knots and unexpected detours.

And here's the thing about that mess—it's not a bug. It's not a sign you're doing it wrong. It's not evidence that you're bad with money or that you lack discipline. The mess is the point. The mess is what makes us human.

The Seduction of the Straight Line

There's something deeply appealing about algorithmic advice. It’s so clean. Plug in your numbers, and out comes a plan: no ambiguity, no second-guessing. Just follow the formula.

When you're overwhelmed by financial decisions, a straight line feels like relief. Someone—or something—finally has the answer. “Just tell me what to do, and I'll do it!”

Everything’s mapped out. It’s paint by numbers, just like when you were a kid.

But here's what the algorithm doesn't know: it doesn't know that your mother just got diagnosed with cancer and you're trying to figure out if you can afford to take unpaid leave. It doesn't know that your child is struggling in school and needs a tutor you hadn't budgeted for. It doesn't know that you just got an unexpected bonus and you're torn between paying down debt, investing, or finally taking that trip you've been postponing for five years.

The algorithm doesn't know that you're human, and life changes.

Why Math Isn't Enough

Don’t be mistaken - the maths matters. Of course it does! Compound interest is real. Time value of money is real. The difference between a 6% return and an 8% return over thirty years is very real.

But when we reduce money to maths alone, we forget what it feels like to make decisions when you're scared. Or uncertain. Or grieving. Or excited. Or exhausted. Or newly in love. Or watching your industry collapse. Or getting a second chance you never expected.

Financial decisions aren't made in a vacuum. They're made in the tangled middle of actual lives.

That's why human financial advice still matters. Not because humans are better at maths than machines—we're definitely not. But because good advisors know that the maths is just the beginning. The real work is helping people navigate the gap between what the spreadsheet says they should do and what feels possible in their actual circumstances.

Algorithms Optimise, Humans Navigate

Here's what I've learned after years of working with people and their money: algorithms optimise for efficiency. Humans navigate complexity.

An algorithm can tell you the mathematically optimal move. But it can't tell you whether that move is worth the fight it'll cause with your spouse. It can't weigh the emotional cost of saying no to your child’s sports travel team against the financial benefit of staying on track. It can't factor in the value of sleeping soundly at night, even if that means choosing a less "optimal" investment.

There's a reason Japanese retirement homes started removing robots and bringing back human caregivers.1 The robots were more efficient. They didn't get tired. They didn't call in sick. They could lift residents without risking back injuries. But the residents wanted the human touch. They wanted someone who could sense when they needed comfort, not just assistance. Someone who could respond to mood, not just medication schedules.

The same principle applies to money. The algorithm gives you the straight line. The human advisor helps you draw your actual path through the tangled mess.

And sometimes the best financial decision isn't the one that maximizes your net worth. Sometimes it's the one that lets you live with yourself. Sometimes it's the one that honours your values, even when it costs you. Sometimes it's the one that acknowledges you're not just a rational economic agent making optimal choices—you're a person trying to build a life that matters.

You Don't Know Where You Sit on the Curve

Late last year, I wrote about how no one actually knows where they sit on the curve of life's probabilities.2 The algorithm assumes average. But you're not living an average life—you're living your specific life, with your specific luck (good and bad) in any given year. My claims year proved that perfectly.

Some years, you sail through with nothing but routine expenses. The algorithm would call that "optimal."

Other years, everything hits at once. Three family emergencies, a job loss, a health scare, and a busted gearbox. The algorithm would call that "suboptimal" or "poor planning."

Yet, both years are just… life. You didn't do anything wrong in the hard year. You didn't do anything especially right in the easy year. You just lived as normal, where probability meets reality and the straight line becomes a scribble.

The Question Worth Asking

So here's what I want you to ask someone you care about today: What did your budget not account for this past year?

Budgets are great. I believe in them. But they're not magic. Real life always sneaks something in. The car repair. The friend's wedding at the other end of the country. The opportunity you couldn't pass up. The emergency that wasn't really an emergency but felt like one at the time.

Those deviations from the plan? They're not failures. They're data. They're information about what your life actually requires, not what the algorithm thinks it should require.

The straight line is beautiful. But the tangled mess is real. And real is where we have to learn to make good decisions.

Why We Still Seek Human Advice

Here's the deeper truth about why people still seek human financial advice in an age of robo-advisors and AI-powered planning tools: life is dynamic, and our responses need to be too.

A good financial plan isn't static. It breathes. It adapts. It changes when your circumstances change, when your values shift, when unexpected opportunities arise or unwanted challenges appear.

The algorithm updates when you feed it new numbers. The human advisor updates when they see the worry in your eyes, hear the excitement in your voice, sense the hesitation you can't quite articulate. They adjust not just to what has changed, but to how you've changed.

Because here's what the tangled mess really represents: not chaos, but adaptation. Not failure, but responsiveness. Not a deviation from the plan, but evidence that you're paying attention to your actual life and adjusting accordingly.

The straight line assumes the future will be like the past. The tangled line knows better. It knows that life zigs when you expect it to zag. It knows that the best plan is one that can bend without breaking, that can accommodate both disaster and delight, that can hold space for the full complexity of being human.

That's not a bug in the system. That's the whole point of having a life worth planning for.

Nick Stewart

(Ngāi Tahu, Ngāti Huirapa, Ngāti Māmoe, Ngāti Waitaha)

Financial Adviser and CEO at Stewart Group

• Stewart Group is a Hawke's Bay and Wellington based CEFEX & BCorp certified financial planning and advisory firm providing personal fiduciary services, Wealth Management, Risk Insurance & KiwiSaver scheme solutions.

• The information provided, or any opinions expressed in this article, are of a general nature only and should not be construed or relied on as a recommendation to invest in a financial product or class of financial products. You should seek financial advice specific to your circumstances from a Financial Adviser before making any financial decisions. A disclosure statement can be obtained free of charge by calling 0800 878 961 or visit our website, www.stewartgroup.co.nz

• Article no. 440

---

References

1. James Wright's research on Japanese eldercare facilities found that care workers often rejected robots like the "Hug" lifting device, preferring to care with their own hands and finding it more respectful to residents. See: Wright, James. Robots Won't Save Japan: An Ethnography of Eldercare Automation (Cornell University Press, 2023); and MIT Technology Review's coverage of robot implementation challenges in Japanese care homes (January 2023). ↩

2. "Why Self-Insurance Rarely Works," Stewart Group, December 5, 2024 ↩

Leaving Home?

Everybody loves the comforts of home, but investors who become too anchored to familiar territory can end up with a very narrow view of the world.

Home bias, the tendency of investors to allocate a disproportionate amount of their funds to their domestic market, is a well-documented phenomenon.

As at 31 December 2016, superannuation funds regulated by the Australian Prudential Regulation Authority had nearly 50% of their total equity exposure in local shares. For self-managed super funds, the average home allocation was 72%.

There can be rational reasons for home bias. Australian investors, for instance, have the advantage of dividend imputation. This is where firms that have already paid income tax on profits attach tax credits in distributions to shareholders.

A second understandable reason they might tilt to their home market is familiarity with the companies they are investing in. Companies like Westpac, ANZ, Qantas and Air New Zealand are household names and are frequently in the news.

A third justification might be that a home bias satisfies an individual investor’s particular goals or risk appetite. These are all issues for an advisor to consider.

But a large home bias also can have undesirable consequences. Those consequences relate to the risks of a portfolio ending up with very concentrated exposure to individual countries, companies, and sectors.

DIVIDING UP THE WORLD

This is particularly the case where your home market is relatively small in a global sense or where it is dominated by one or two sectors.

Chart 1 shows the natural weights of some of the countries in the MSCI All Country World index, a popular benchmark for the global share market. You can see the US is by far the biggest market, with a weight of more than 53% as at 31 January 2017.

Japan is a distant second with a weight just above 8%, then the UK, Canada, France and Germany, China and Switzerland. Australia is ninth with a weight of 2.4%, while New Zealand is in 36th place with a weight of just 0.1%, too small to see on the chart.

Now, what happens when we bias our portfolio to the home market? Chart 2 shows the country representation in an equity portfolio with a home bias of 60% to Australia.

As a result of this tilt, we have to scale down representation of other major markets. For instance, the US exposure is lowered from 53% to 22% and emerging markets from 11% to just 4%. Put another way, the weight of all countries, apart from Australia and the US, in this portfolio is just 18%.

So you can see that this degree of home bias represents a pronounced deviation from the global market portfolio and leaves you taking unnecessarily big bets. For instance, Canada was the best performing developed market in 2016 with a gain of more than 25%. But with our home bias, the weight to Canada is reduced by two thirds.

Since we have no reliable way of predicting which country will be the best performer year of year, it makes sense not to deviate too far from the market portfolio.

DIVIDING UP THE SECTORS

Having a sizeable home bias means not only taking a big bet on a single country, but ending up with a disproportionate exposure to certain sectors.

In Australia’s case, the big four banks represented nearly 28% of the market, as of early 2017. Financials overall made up 38%, while materials stocks represented 17%.  In other words, two sectors made up more than half of the market.

To put that into perspective, financials represented just 19% of the global market while materials had a share of just 6%.

Another way of looking at this bias is to consider which sectors are poorly represented in Australia. For instance, information technology stocks made up just 1.3% of the local market, compared to 16% globally.

Chart 3 shows the natural market weight of sectors in a global portfolio. Chart 4 shows what happens when you have a 60% home bias. In other words, you end up taking a significant bet on financials and materials stocks, at the expense of other sectors like technology and healthcare. It’s a much less diversified portfolio.

DIVIDING UP THE STOCKS

We’ve seen that a home bias means concentrated bets on a single country and a couple of sectors. So it shouldn’t be a surprise that it also means you end up taking outsized bets on a handful of stocks.

With a 60% Australian home bias, a single stock—the Commonwealth Bank—will represent nearly 6% of your portfolio. That is more than the weight of the UK and Japan combined or more than your entire holdings in emerging markets.

Put another way, the top five stocks (the four big Aussie banks and BHP Billiton) will make up more than 20% of your portfolio. You will also more than halve your exposure to big overseas names like Apple, Microsoft, Amazon, Vodafone, Nestle and Volvo.

In fact, of the top 50 stocks in this Australian-biased “global” portfolio, all but four are local stocks, the exceptions being Apple, Microsoft, Exxon Mobil and Amazon.

Chart 5 illustrates how concentrated this portfolio becomes once the home bias is accounted for.

PUTTING IT ALL TOGETHER

Given investors tend to source most of their income from their home nation and hold most of their other assets there, you can see that this degree of home bias represents a very big bet on one country, a couple of sectors and a handful of stocks.

So the question then becomes what degree of home bias is acceptable. It shouldn’t be surprising that there is no one right answer to that. It depends on each individual investor’s tastes, preferences, circumstances and goals.

A good approach is to use a global market portfolio as your starting point. If you want to increase the expected return of your portfolio, you can use information in current market prices and company fundamentals to tilt your portfolio towards stocks with higher expected returns. Research has shown that small caps, low relative price and high profitability stocks deliver a premium over the market return.

While you may tilt your portfolio towards Australia for the franking credits or your familiarity with the stocks, it’s important to understand this has nothing to do with the expected return of your portfolio. It is just a preference. But it also comes at a cost.

In contrast, broad global diversification creates a portfolio that spreads its risk to more economies, to a greater number of stocks, to a wider range of companies and to a wider spread of sectors.

Again, there is no single right answer in terms of asset allocation. It will depend on the individual investor’s circumstances, goals and risk appetite.

But it is worth reflecting on the fact that an investor whose allocation is made up 60 percent of Australian equities is overweighting Australia relative to the global market by a factor of 25.

At some time or another, we all have to leave home.

• Source: Jim Parker, Vice President, DFA Australia Limited

In Long Run, There’s No Such Thing as an Einstein Investor

There are no easy answers in investing. It is tempting to replicate a successful strategy — one created by an outstanding investor, like Warren Buffett, or through in-depth statistical analysis of the wisdom of crowds — and such approaches can actually work for long periods.

But paint-by-number portfolios won’t succeed forever. And without deep expertise, it makes little sense to veer much from a simple market portfolio — one that seeks to match the overall performance of the market, and not beat it.

These reflections are prompted by the television series “Genius” (based on the Walter Isaacson biography “Einstein: His Life and Universe”), which I’ve been watching on National Geographic TV. The series also inspired me to reread Einstein’s own popularization of his theories, in the book “Relativity: The Special and General Theory.”

Albert Einstein, who may have been the most famous person ever to be publicly identified as a genius, had a disrespectful attitude toward the dignitaries of the physics profession of his time, and a lonely and unique approach to science.

Yet great as Einstein’s theories were, others in the scientific community had been on the verge of discovering them when he came along. In fact, it is possible to argue that large numbers of collegial scholars who do not keep secrets, do not pretend to know everything and share freely will eventually surpass the achievements of a lone genius.

Similar debates dominate professional investing. For help in making important financial decisions, some of us are looking for Einsteins, others for communities of scholars or professional money managers with solid ideas.

In terms of popular reputation, Mr. Buffett may be investing’s closest approximation of an Einstein. Some investors have done well simply by copying Mr. Buffett’s financial moves.

On the other hand, many investors embrace the catchy methods that bubble up from time to time, like “smart beta,” a phrase for a form of systematic investing that claims to outperform the market. There is no universal agreement on what smart beta means, but it typically refers to published theories and replicable statistical analysis, and mechanical procedures aimed at beating them. Smart beta clearly is the epitome of community property, not a quirky genius.

A lone investor, whether a genius or not, can typically keep a secret better than a community can, and does not have to publish his methods. This difference is important but not absolute because the Securities and Exchange Commission generally requires large institutional investment managers to file quarterly reports listing their holdings.

Through the years, despite Mr. Buffett’s admonitions not to do so, many people have tried to mirror his strategy as it is revealed on the Berkshire Hathaway Form 13F filed with the S.E.C. Because Mr. Buffett presents himself as a long-term value investor, investors may think it doesn’t matter that this filing may be months out of date. But not understanding exactly how he makes decisions, they don’t have his edge, must come to the party late and have frequently bid up prices as they compete against one another to buy the assets in his portfolio.

There is plenty of evidence of this: A 2008 study by Gerald S. Martin of American University and John Puthenpurackal of the University of Nevada, Las Vegas, found that when S.E.C. filings reveal changes in the Berkshire Hathaway portfolio, the stock prices of newly acquired companies had an abnormal one-day increase averaging 4 percent. Even so, they found long-lasting effects. A simulated replication strategy from 1976 through 2006 based on the S.E.C. filings outperformed the market by over 10 percent a year.

That was an amazing result, though merely copying Mr. Buffett has been less satisfying in recent years because his investment performance has dimmed somewhat. No one can excel all the time, and even a Buffett may produce in a lifetime no more than a few great ideas that may not be viable forever.

There are even bigger problems in replicating strategies extracted from the community of scholars who publish not only what they do, but why they do it.

For example, a much-talked-about paper by R. David McLean of Georgetown University and Jeffrey Pontiff of Boston College published last year pointed out that the effectiveness of stock market investing strategies seems to diminish, but not disappear, after publication.

The paper, which won the American Finance Association’s 2016 Amundi Smith Breeden Award, examined 97 financial patterns that appeared to predict investment returns and had been published in reputable scholarly journals and supported by tests that found statistical significance. Such strategies relied on factors like price-earnings ratios, changes in analyst recommendations, credit rating downgrades, stock price momentum, industry momentum and failure to pay dividends.

The researchers looked at the performance of each of these strategies, assuming you had started right after the publication of research papers on them and then continued for years. They found that while the strategies outperformed the market, their success decreased by more than 50 percent after publication.

In a follow-up paper, the two authors, along with Joseph Engelberg of the University of California, San Diego, showed that one-day positive surprises on firms’ earnings announcements accounted for nearly all of the investment’s total outperformance. Why? It appears to be because the market consistently makes mistaken valuations of corporate earnings, which tend to be corrected in stock prices only when the final earnings evidence is staring traders in the face.

So what’s an investor to do? Both published statistical analyses and published actions and opinions of knowledgeable people, whether geniuses or just smart and well-informed investors, are worth mulling over if you have a taste for such things. But don’t follow these strategies blindly. We need to exercise our intuitive judgment as well as rely on the wisdom of smart, well-informed people to decide whether to continue to rely on statistical indicators and investment strategies that seemed to work in the past.

The problem is that the world is too complex for any method to work all of the time. The economist Alfred Marshall, then of Cambridge University, wrote in his 1890 textbook “Principles of Economics”: “Although scientific machinery should be as definite as possible, at the same time it should be flexible.” He added, “There is so much variety in economic problems, economic causes are intermingled with others in so many different ways, that exact scientific reasoning will seldom bring us all the way to the conclusion for which we are seeking.”

His reasoning is still valid. We need to use statistical analysis but also respect human intuition and even genius if we are able to identify it. But do so with caution. No single strategy is likely to beat the market forever.

• Source: NY Times