Devina Mehra: Are you driving your investment portfolio by looking at the rearview mirror?

You know what will happen if you drive looking at the road that is already behind you? Ignoring both the dangers and opportunities ahead? That is what I thought of when I saw that inflows into Indian gold exchange traded funds (ETFs) exceeded flows into equity funds in January 2026.

For context, in March and April 2025, gold inflows were hugging the zero line even as equity mutual fund flows were at pretty much the same level as in January—around ₹24,000-25,000 crore. Now the dollar price of gold is down 20-25% from its January peak.

But this is not a column about gold. I have already written about gold prices earlier explaining that this metal has been historically a more volatile asset than equities and its 1980 high was not crossed for 27 long years. The general point here is about chasing returns in whatever has done well of late, whether it is an asset class, country, sector or market cap category or even an investment strategy.

Just as you will crash your car if you drive looking backwards, data shows that if you systematically invest in themes, asset classes or geographies that have done well in the previous period, you will equally systematically underperform. However alluring it looks, planning your future path based on the road behind you simply does not work.

Now something closer home. Most investors invest in a particular mutual fund or portfolio management services (PMS) scheme based on recent performance. Let us see how well this strategy works.

DSP, in its January issue of Netra, collated data on mutual fund schemes in the top quartile (the top 25%) in their category for 3 years and found that 60-100% of them were not in the top quartile in the succeeding 3 years. They did it across eight time periods and five mutual fund categories and these were the findings. On average, over 80% of the top performing schemes over a 3-year period are not in the top 25% category in the next 3 years.

This holds across markets. Investors chase recent performance, which, if anything, is negatively correlated with the next period’s performance.

Remember Peter Lynch whose book One up on Wall Street is considered an investment classic. He is the legendary Fidelity fund manager who delivered 29% compounded growth for 14 years in the Magellan fund. However, as per Fidelity’s own data, the average investor in the fund compounded only 7% and many lost money.

Lost money in the best performing mutual fund in US history! Why?

Because investors entered and exited at the wrong time. For example, the fund gave a 20% return in 1980, attracting a slew of new investors, but then substantially underperformed in the following year, prompting many to exit. If they had stayed invested, they would have seen some spectacular returns over the next decade.

In our own equity PMS scheme, we have seen the same pattern. While over a 6-year period the scheme has outperformed benchmarks substantially, it has not done so every year. It is almost an alternate-year pattern, with some underperformance in a year and significant outperformance in the following year. That is the way it is supposed to be, but try telling it to restless investors.

Sebi data shows that most mutual fund investors do not remain invested in a mutual fund scheme for even two years at a stretch. All the compounding seen in beautiful Excel spreadsheets is almost never experienced by individual investors, thanks to their own actions.

It is the same story with investment themes. Whether it is small cap, defence IT, PSUs or whatever, investors get into a theme that has lately done well only to find that they’ve entered at a peak. A smart strategy is not to get onto a train that has long left the station... Wait for the next one!

Increasingly, I find professionals are no better. All around me are ‘experts’ who pontificate on blinding glimpses of the obvious. Gems include ‘India is an expensive market because its P/E is higher than that of many other markets,’ ‘foreign portfolio investors have been net sellers in India’ (who knew!) and ‘India has no AI story and is not an innovation market.’All this is over and done with. What matters is what happens now.

Consider the US Magnificent Seven tech majors: Microsoft, Google, Nvidia, Meta, Apple Amazon and Tesla. They contributed 60%-plus of the 2023 move in the S&P 500, over 50% in 2024 and more than 40% in 2025. In 2025, only two, Google and Nvidia, drove the move while the other five underperformed the index. In the first six weeks of 2026, all seven were down, even as stocks like Exxon, Chevron, Walmart and Johnson & Johnson rose. The AI bull is tiring.

As an aside, there are other danger signs in AI like an estimated capital expenditure of about $700 billion by the four AI companies—about 4.5 times their capex in 2023. Whether this will ever make adequate return on capital is anyone’s guess. Given that data centres are extremely energy and water intensive, the West Asia war may push a lot of AI plans further down the road.

In India, what matters is the earnings trajectory from here on and valuation ratios. Market commentary on moves that have already happened doesn’t help. Markets are about both the pitfalls and opportunities that lie ahead, not what we have already passed!

The author is founder of First Global and author of ‘Money, Myths and Mantras: The Ultimate Investment Guide’. Her X handle is @devinamehra