Mr Scholes, who won a Nobel Prize in economics in 1997 and co-founded failed hedge fund Long Term Capital Management, spoke about the time diversification at the UNSW International Business Forum in Sydney this week.
A major failing of the financial services sector is its assumption that risks are constant over time, said Mr Scholes.
"In financial services we pretend that there is not a changing distribution of returns over time."
Focusing on average returns is a mistake, he said, because "it's really compound returns that count".
"Time diversification is trying to have a target volatility equal to your actual volatility experienced.
"In managing money over time we have to think about keeping risk at target or as close to target as possible," he said.
Over the past 15 years, the focus in the finance industry has been concentrated on relative performance at the expense of compound return (or 'absolute' performance), Mr Scholes said.
"So managers are selected if they can add average [returns] which exceed the return on a benchmark. The problem is that the benchmark risk could be changing dramatically."
In addition, asset owners tend to select fund managers relative to a benchmark because doing so "protects them from their boards who are worried about their performance", Mr Scholes said.
"When investors are required to manage money to stay close to a benchmark, then they tend to be concentrated more heavily in higher risk stocks.
"If we are focusing on averages, using historical data, we’re assuming the risks are constant – and they’re not," Mr Scholes said.
