Financial planners who recommend dollar-cost averaging over lump-sum investing have often been at odds with academics, who over the decades have pointed to study after study indicating that lump-sum investing produces better returns. But that may now change.

A new study published last week by Thomas O’Brien, a professor at University of Connecticut’s Department of Finance, showed that there are indeed scenarios where dollar-cost averaging (DCA), where money is invested gradually instead of all at once, is better than lump-sum investing.

According to O’Brien, there was a long period of time where the research simply condemned DCA, but from 2000 to 2015 researchers tried to understand why the practice has remained popular with advisors in the face of academic counter-evidence.

This is what has been referred to as the “DCA puzzle,” and this is what O’Brien said he wanted to examine in Dollar Cost Averaging vs Optimal Buy-And-Hold With Equity Momentum and Reversals.

“Look, I don’t have a dog in the hunt on whether financial advisors continue to advise DCA. I don’t have a problem with them doing it, and I don’t have a problem with them not doing it,” O’Brien said. “But the academic people have been saying it’s no good. And what I’m saying is, it’s worth another look.”

The study’s framework involved a model that examined returns over three years, where the first year registered upward momentum on equity prices followed by a two-year reversal. The beginning equity price was normalized to one dollar, and fixed-income securities yielded a constant risk-free rate of 2% per half year.

When modeling the momentum phase, the study differentiated between fair-value equity pricing and “overreaction” pricing, such as one would see among ordinary, non-professional investors chasing a trend. Then in the reversal phase, the reversal of the overreaction pricing would bring that pricing level back in line with the fair-value equity pricing.

The decision of allocation strategy employed by the model is one financial planners most likely will instantly recognize. “Investors are assumed to face a decision on how to allocate a lump sum of funds. Perhaps the sum is from an inheritance or the sale of a business,” the study explained. “The investors choose between only a simple buy-and-hold strategy and DCA.”

With buy-and-hold, the total amount of the lump sum was invested in a blend of equity and fixed income positions for the entire three-year period, at different levels of risk tolerance. With DCA, the lump-sum was split into three equal installments, which were invested in equities. While the uninvested portions were waiting for investment, they were held in a fixed-income instrument. In addition, the DCA portion was divided into two groups—those that did not rebalance to their risk tolerance every six months and those that did.

At the end of the three years, the lump-sum strategy outperformed dollar-cost averaging without rebalancing for all levels of risk tolerance. However, when investors rebalanced, DCA beat out lump-sum investors for all except the most risk-averse.

The real-world application of the study results is unclear, the study said, as the success of DCA within in the study was dependent on the market conditions of short-run momentum followed by a reversal that was twice as long, as in the model.

“I don’t think there will be a piece of research that will come out in my lifetime that will be the definitive answer to this question,” O’Brien added. “But what we’re saying is there are situations where DCA outperforms lump-sum, and that’s a significant step forward in the academic literature.”