Fundie vs AI: One year on, who is the winner?

A year ago, I set about the half-serious, half-fun task of finding out whether AI was any good at picking stocks. 

The premise and rules of the thought experiment were laid out in the following wire:

Equities

Fundie vs AI: Who is the better stock picker? (and 10 ASX stock ideas for your trouble)

View

Well, this Wednesday saw the year-long experiment for the income portfolios conclude. So now, we get to dive into the delicious results. I won't keep you in suspense. 

It pleases me greatly to report that, despite all of the advances in AI, it still doesn't match human stock pickers - or at least it didn't one year ago, based on this hardly exhaustive experiment. Still, one for the humans, I would have thought. 

The losing AI portfolio

As can be seen from the table below, the AI-generated portfolio delivered the following; 

• A total return of 22.03%
• A dividend yield of 4.10%

The winning portfolio from Plato's Dr Don Hamson

As can be seen from the table below, Dr Hamson's human-generated portfolio delivered the following:

• A total return of 22.39% (vs. 22.03% for AI)
• A dividend yield of 6.08% (vs. 4.10% for AI)

Some of you might consider that to be a close result. I would be more likely to call it 'resounding'. The aim of the portfolio was to deliver income, and a 1.98% positive difference offered by Dr Hamson's portfolio is considerable. Not to mention the fact that the total return is also greater, which is no mean feat in itself. We're not talking about growth stocks here, where the dispersion of returns would be much wider. 

Finally, Dr Hamson also made a point of highlighting that he and his team focus on stocks with franking credits, and the returns would have been even greater if they were taken into account. 

The aftermath

We asked Bard, the AI used to pick stocks a year ago, whether it would make the same choices based on the current market environment. Here is the prompt we fed into it and the results it generated. 

Prompt:

• Harvey Norman (ASX: HVN): High dividend yield & potential rebound after a profit dip.
• National Storage REIT (ASX: NSR): Growing demand for self-storage & consistent dividends.
• Transurban (ASX: TCL): Toll giant benefiting from increased traffic & a reliable income stream.
• Macquarie Group (ASX: MQG): Diversified financial powerhouse with a solid dividend history.
• Dexus Property Group (ASX: DXS): Office landlord with a recovering rental market & an attractive yield.

Given the changes in the market, earnings, US election, economic forecast changes, the commencement of rate cuts and outlook over the past year, would you still hold these same stocks? Let me know if you've chosen to remove any, and which ASX stock replaces it.

Reply:

National Storage: I would HOLD National Storage REIT. The demand for self-storage continues to grow, driven by factors like downsizing, lifestyle changes, and business needs. NSR has a strong portfolio of properties and a history of consistent dividend payouts. While interest rate rises might impact borrowing costs, the underlying demand for their services provides stability.

Transurban: I would HOLD Transurban. Toll roads are essential infrastructure, and TCL benefits from increasing traffic volumes as cities grow. Their long-term contracts often include inflation-linked toll increases, providing some protection against rising costs. While economic conditions can impact traffic, TCL's essential service nature and dividend history make it a reliable income stock.

Macquarie Group: I would HOLD Macquarie Group. MQG is a diversified financial powerhouse with a global presence. While financial markets can be volatile, MQG has a strong track record of navigating different economic cycles. Their diversified business model, including investment banking, funds management, and commodities, provides resilience. The dividend yield is attractive, and they have a history of growing dividends over time.

Dexus Property Group: I would REPLACE Dexus Property Group with Stockland (ASX: SGP). While DXS is a solid company, the office property market is facing some headwinds with the rise of remote work. SGP offers a more diversified exposure to the property sector, with a mix of retail, logistics, and residential developments. This diversification can provide more stability in a changing economic environment. SGP also has a strong track record of dividend payments and offers an attractive yield.

Reason for the change:

• Diversification: In the current environment, diversification is key. SGP's mix of property assets provides a broader exposure to the market, reducing reliance on any single sector like office space.
• Resilience: SGP's focus on essential retail and logistics properties can provide more stability during economic uncertainty.
• Dividend history: SGP has a consistent history of paying dividends and offers an attractive yield, making it a suitable replacement for DXS in an income-focused portfolio.

This is just my [Bard's] opinion based on the information available. It's crucial to do your own research and consider your individual investment goals and risk tolerance before making any investment decisions. The market is constantly changing, and past performance is not indicative of future results.

The last word from the Doctor

Dr Hamson also provided an update on his original picks, which were: 

• Ampol (ASX: ALD)
• JB Hi-Fi (ASX: JBH)
• Macquarie (ASX: MQG)
• Medibank Private (ASX: MPL)
• Rio Tinto (ASX: RIO) 

"We still like JBH and RIO, so no real change there.

Ampol disappointed with a profit downgrade prior to announcing its earnings in August. Lower refining margins and higher interest costs impacted earnings. Given declining earnings, it is less likely to pay out excess franking in the near term.

Macquarie’s earnings also disappointed, with returns from their green investments yet to flow through.

For Medibank, whilst we are still positive about insurers in general, we now prefer general insurers".

It seems AI still has some way to go when it comes to picking income stocks... and brevity, for that matter.