Dollar-Cost Averaging: The Trade-Off Between Risk and Return
Journal of Financial Planning
ISSN or ISBN
- Despite lump-sum investing’s superior performance, dollar-cost averaging continues to be popular among investors.
- Much of the previous literature has used behavioral reasoning, seasonality, conditional expected returns, and mean reversion in stock prices to explain this popularity.
- Using a standard mean-variance analysis, this paper finds that dollar-cost averaging can be used to lower risk.
- Results suggest that depending on an investor’s level of risk aversion, dollar-cost averaging can be an optimal investment strategy.
Instead of investing money in one lump-sum, dollar-cost averaging (DCA) is a strategy where an investor invests money into risky assets by allocating equal dollar amounts over a period of time. Practitioners have championed this investment method for a long time, arguing that DCA minimizes losses in a portfolio’s value if the price of a risky asset gradually declines. Lump-sum investing is thus considered risky, as the price of a risky asset can immediately decline after investing. Proponents of DCA also claim that it can provide a lower average purchase price for a risky asset, suggesting that purchases of more risky assets with DCA when prices are declining will provide better returns than lump-sum investing.
Despite the fact that the majority of academic research has shown that DCA is inferior to lump-sum investing (as well as other investment strategies), the popularity of DCA remains high among practitioners and the investing public.
Previous studies have sought to explain the popularity of DCA via behavioral finance. Statman (1995) attributed the use of DCA to investor irrationality, proposing four behavioral explanations for the popularity of DCA: (1) prospect theory, which posits that people tend to give less weight to outcomes that are merely probable than to outcomes that can be obtained with certainty; (2) aversion to responsibility and regret; (3) cognitive errors caused by recent stock trends; and (4) a lack of self-control.
Leggio and Lien (2001) empirically tested Statman’s behavioral rationale for the persistence of DCA. However, their findings contradicted Statman (1995). They found that lump-sum investing strategies were better than DCA approaches. They suggested that prospect theory preferences and loss aversion alone could not explain the popularity of DCA strategies. They also noted that DCA strategies perform worse when used to purchase more volatile stocks—an interesting finding considering that DCA has been purported to perform better for those holding volatile stocks.
Similarly, Dichtl and Drobetz (2011) used Monte Carlo and bootstrap simulations, which included cumulative prospect theory, to compare DCA strategies with lump-sum and buy-and-hold strategies. They found that the DCA strategy was not mean-variance efficient. They also found the DCA strategy had higher cumulative prospect values than the lump-sum investing strategy, but a 50-50 buy-and-hold strategy had higher cumulative prospect values than either of the other strategies.
Several studies attributed the wide acceptance of DCA to such factors as the seasonality of equity returns, high conditional expected returns of DCA, and mean reversion in stock prices.
Atra and Mann (2001) considered the seasonality of equity returns when they compared the performances of a DCA strategy with a lump-sum investing strategy. They found that when initiated from October to January, the lump-sum investing strategy outperformed DCA, but DCA outperformed the lump-sum investing strategy when initiated from February to September.
Milevsky and Posner (2003) used stochastic calculus and Brownian bridges to show that if an investor knows the final value of a security, and if the stock is highly volatile, then the conditional expected return from a dollar-cost average investment will exceed the expected return from a lump-sum investment for the same security.
Trainor (2005) showed that the probability of losing a particular amount at any given time within an investment horizon is significantly reduced when investors use DCA strategies instead of lump-sum investing strategies.
Brennan, Li, and Torous (2005) compared certainty equivalents between DCA and lump-sum investing strategies, concluding that mean reversion in stock prices from 1926 to 2003 may have benefited investors using DCA strategies, especially if those investors were adding new stocks to an already well-diversified portfolio.
Cho, David D. and Kuvvet, Emre, "Dollar-Cost Averaging: The Trade-Off Between Risk and Return" (2015). HCBE Faculty Articles. 769.