Investors must reject outdated models and assumptions and take a more realistic view of the world, argued Amlan Roy at last week’s IPE Conference.

Roy, a specialist in macro-demographics and senior research associate at the London School of Economics, urged delegates to “question and challenge the basic concepts” that underpin investment models.

“We have had CEOs and CIOs today talk about [model portfolio theory] and the capital asset pricing model,” he said. “It’s a single-period model, and it fails the real life test.

“We are pressing [refresh] on models that are single-period, developed in the 60s, and people are still swallowing it.”

Such models are usually based on a portfolio of equities, bonds and cash, Roy said, which ignores the recent rise of allocations to alternative asset classes such as real estate, infrastructure and private equity.

“For the last 5-10 years, we’ve been saying equities and bonds cannot meet your liabilities,” Roy said. “They will not give you the kind of returns you bought them for in the 1970s and 1980s.”

Roy also said investors and pension fund managers should consider how macroeconomics and demographics were “influencing the fundamentals of both assets and liabilities”, while avoiding long-held assumptions of what constitutes important data.

For example, on the assets side, he argued that the size of a population or simple GDP were insufficient indicators of a strong economy.

Instead, investors should look at GDP per capita as an indication of growing potential for a population’s disposable income.

On liabilities, Roy used several examples from around the world to illustrate how retired populations were changing more than many had predicted.

Globally, the population aged 80 years and above grew by 400% between 1970 and 2015, he said, “yet the world population only grew by 100%”.

He added: “Japan’s 80-plus population was 1% in the 1970s, and 8% in 2015. The super-old population is not just growing faster, they are much older.”

Finally, Roy issued a stark warning that “no country in the world has enough money to pay for these pensions”.

“Please don’t use models because they behave badly – you need to understand limitations and ask the right types of questions,” he concluded.