Time to dream with long-term shares, artificial intelligence

It is February already and, for most of us, our relaxing holiday experiences now seem like a distant memory.

However, there is always the ability to dream and save for the next holiday. Among my holiday reading there were two experiences that have stayed with me; a business book analysing history and a piece of software, which according to some, will reshape the future.

The book was Stocks for the Long Run by Jeremy Siegel. In our language, the title translates to "shares for the long run". The book considers the performance of different asset classes through history.

Given Siegel’s background as an economics professor at The Wharton School of the University of Pennsylvania you could be mistaken for thinking it would be heavy going. Although it does require a degree of investment literacy, it is fascinating as it reviews the last 220 years of investment market experience.

It compares the relative performance of gold, term deposits, bonds and shares over that period. It concludes that shares have the ability to generate an average compounding return (above inflation) of 6.9% a year.

This performance makes shares superior to all other asset classes. Their out-performance was significant, being almost twice the return of bonds which were the second-best performing asset class at 3.6% a year.

Another surprise was the performance of the supposed safe haven asset of gold. Over that period gold showed an average return of less than 1% per annum (above inflation).

The book also considers the risk associated with each asset class. As one might expect, over the short term (i.e. 12 months) shares showed the greatest variation of return (i.e. volatility). However, on average, you only had to hold shares for 10 years before the average level of volatility reduced below that of bonds.

The reality is, over time shares are superior to bonds on both a return and risk basis. This would be a surprise to many New Zealanders. Given the fact that most New Zealanders will be exposed to investment markets for more than 30 years through their KiwiSaver portfolios, it makes one think hard about the most appropriate investment mix. When it comes to returns; clearly the more shares the better.

The other interesting finding was that the return provided by shares was remarkably stable through time. Whether it was the period from 1801 to 1870, when America transitioned from an agrarian to an industrial economy and could be thought of as comparable with an emerging market of today.

Or from 1871 until 1925, when the United States economy was transformed by the internal combustion engine and became the foremost world economic power. Or more latterly from 1926 until now, a period that covers the Great Depression, World War 2, the Tech Crash, the Global Financial Crisis and the 2020 Covid pandemic. Through each period in history the return was remarkably consistent — around 6.9% per annum (after inflation).

There has already been quite a lot of media comment about my second discovery, ChatGTP. For those who may not have heard of ChatGTP, it is an artificial intelligence (AI) tool which allows you to generate conversational text on any question you might have.

I was given a demonstration of this technology by one of my millennial sons; and boy is it impressive. As a test, I asked it to write 500 words on the challenge of developing a financial plan for a couple with different risk profiles.

In five seconds, it created a coherent piece of writing that covered the main points that I would have raised if such a question was asked of me. The content was excellent but what was truly impressive was the ability to modify the style and tone to make it more approachable.

We started off with a normal tone, then transformed the text to as spoken by a pirate, then finally to reflect the style of a privileged baby boomer (my son’s suggestion).

In each case it had precisely the same key concepts, but the cleverness of the word selection and the nuance of tone was amazing. I can see why educational organisations are finding this technology challenging. This is certainly going to be a new frontier in communication and learning. If we thought that AI was just going to replace manual workers, it is time to think again.

My days as a columnist could well be numbered. Thinking about how it is going to change the world and be monetised just made my head spin — and ChatGTP could not provide a very good answer either!

However, if Prof Siegel is correct, the one safe bet is AI will create new investment opportunities and the return from the share market in which those companies will be represented is likely to be, on average, 6.9% per annum (after inflation).The more things change the more some things stay the same. Over time shares are likely to continue to be the leading asset class.

 - Peter Ashworth is a principal of New Zealand Funds Management Ltd, and is a Dunedin-based financial adviser. The opinions expressed in this column are his own and not necessarily that of his employer. His disclosure statements are available on request and free of charge.