I have a dystopian nightmare where computers make investment decisions and robots trade securities, while hordes of ragged former portfolio managers roam the City of London and Wall Street shouting about earnings misses and bid-offer spreads.

A little dramatic maybe, but the industry is facing significant challenges from lower returns and digitalisation. Active managers are under pressure from passive strategies while financial repression has reduced the opportunity set for alpha generation. Things need to change.

However, my less traumatic vision of the future is a new kind of alpha – one that doesn’t promise too much, that incorporates responsible investing, that charges lower fees than in the past but retains the flesh and blood anchor to what is essentially a branch of the social sciences.

The asset management industry does need to embrace technology faster than it has to date, to industrialise and become more efficient. But it also needs to be careful not to throw the baby out with the bath water.

The rise of the machines

The asset management industry is faced with profound challenges. Low financial market returns in the post-financial crisis era have contributed to the general decline in fees for asset managers. Low returns and reduced volatility have made it more difficult to generate traditional alpha through stock picking, asset allocation and macro calls.

Read more: Challenger banks explained: A new breed of banking

This has fuelled the rise in lower-cost passive investing. At the same time digitalisation and the application of artificial intelligence and data science is threatening the traditional fund manager role. There is an underlying belief that advanced quantitative techniques can do a better and cheaper job of exploiting anomalies in market pricing than a human can. Recently I read some arguments suggesting the industry is heading towards a domination of passive investing at almost zero fees. If this is the case, how can the industry survive in its current form?

I firmly believe that active investing won’t disappear – I also strongly feel that a purely passive approach can lead to a misallocation of capital. It’s a well understood argument in the bond market where market-weighted bond indices draw investors into taking passive risks that might not be appropriate.

Capital is allocated to the largest debtors. In recent years, a passive approach in a government bond index would have led to longer and longer duration exposure. In a credit index, the average quality has declined given the rise in BBB-rated bonds in the overall asset mix. At the same time, returns have fallen as yields have trended lower.

In equity markets, I struggle to see how a stock’s valuation can reflect its fundamentals if capital is allocated to it just based on that stock being in an index. If all equity investing was passive, then there’d be no judgemental influence on a stock’s price despite changes in its earnings profile or its valuation.

Read more: Gender diversity in Earnings Calls: Why do men talk so much?

A new active

Active equity investing is no longer about picking the right stocks ahead of earnings announcements – although that clearly remains an important aspect of an equity fund manager’s role – but about trying to harvest longer term risk premiums that are driven by profound changes to the global economy.

In the bond world, there are different risk premiums that can be exploited in fixed income – carry, duration, inflation, and credit risk. Picking the right bond on the rates curve or the right credit in a particular sector may still produce a little bit of alpha but getting the bigger picture allocation right over longer periods of time is where the real alpha generation will come from.

Environmental, social and governance (ESG) factors will also be part of the new alpha as non-financial factors will allow fund managers to differentiate themselves from each other and from passive index followers. This does not need to be expensive either. Industrialisation of back-office processing, more efficient use of trading platforms, and the more judicious use of research can potentially help reduce the structural cost base in asset management. Of course, regulatory headwinds have not helped in recent years but this only goes to accelerate the changes that would have happened anyway.

Investing still requires instinct and judgement

Arguably, an individual with a data science degree has likely more chance of getting a job in asset management these days than someone with an economics degree. Responsible investment analysts are more in demand than people who can understand a balance sheet and profit and loss account. But isn’t there a risk that there will be loss of the investment theses in all of this?

Capital is allocated based on the best anticipated risk-adjusted returns – this means human knowledge and instinct should continue to play a role. How would a data scientist interpret comments by US Federal Reserve Chair Jerome Powell at a press conference? How would a passive portfolio respond to improvements in macro data?

Read more: How close are we to a cashless society?

As long as decisions at a policy, corporate and consumer level are taken by real people, there is always going to be room for other individuals to make judgements about what those decisions means for rates, credit, and equity valuations.

Computer science may get more and more sophisticated but human behaviour, reflected through political, business, or consumption decisions, is more complex and not always predictable.

Going back to my vision of the future, the utopian version of the future is one where smart people work with smart machines to deliver smart investment outcomes. The asset management industry needs to change to get there – but it’s already moving in the right direction.