COMMENT. Anyone wishing to invest a larger sum is faced with the question of whether to invest all at once or to invest their capital gradually. A historical analysis shows which strategy has proved superior and why investors nevertheless often hesitate.

When a fixed-term deposit matures, a retirement product pays out, a property is sold or an inheritance is received, investors face the question of what to do with the newly available capital. Should they invest the entire amount immediately? Or should they proceed more cautiously and enter the market gradually? After all, prices might soon fall, making it possible to invest at more attractive levels by waiting. Perhaps a market crash is even imminent because the AI bubble is about to burst, oil prices are weighing on the economy or rising interest rates are affecting equity markets.

Many investors choose a gradual approach in such situations, spreading their investment over an extended period. This reduces their dependence on a single entry price. At the same time, they may miss interim market gains because part of their capital remains on the sidelines. This strategy is known as cost averaging, phased investing or, colloquially, the drip-feed investing. The alternative is a lump sum investment of the entire amount.

Lump Sum Investment or Cost Averaging?

In this article, we examine which of the two strategies has performed better historically. We consider an investor who invests an amount in the equity market either immediately in full or in ten equal instalments over ten months. Investment performance is measured using the MSCI World equity index. Initially, we also assume that uninvested capital earns no interest and exclude transaction costs.

If the entire amount is invested immediately, its value develops in line with the MSCI World Index over the following months. Under cost averaging, however, only one tenth of the investment participates in the performance of the MSCI World in the first month, while 90 percent remains uninvested and earns no interest. At the beginning of the second month, another tenth is added, meaning that 20 percent of the total amount now participates in the performance of the MSCI World, while the remaining 80 percent does not. This continues until the final instalment is invested at the beginning of the tenth month.

Once the final instalment has been invested, the performance of cost averaging and the lump sum investment differs only in their respective starting values from that point onward, as both are fully exposed to the MSCI World from the tenth month. To determine which strategy performs better, it is therefore sufficient to compare their performance during the investment phase. We consider all possible investment phases from January 1970 to September 2026, yielding 680 observations.

Waiting Means Forgoing Returns

Historically, the lump sum investment came out ahead in roughly three quarters of all observations. Figure 1 shows that, in monthly comparisons since January 1970, the lump sum investment generated a higher return than cost averaging in 73 percent of cases. Cost averaging performed better in only 27 percent of cases.

Investors who chose cost averaging accepted lower returns on average. Over a ten-month investment period, the average return differential between the strategies was 5.1 percent. The longer the investment period, the greater these opportunity costs for investors.

Figure 2 shows the average return differential as a function of the length of the investment period. Spreading an investment over just two months already comes at a return cost, albeit only 0.5 percent. Over five months, the differential rises to 2.2 percent, and over ten months to the aforementioned 5.1 percent. For a period of two years, or 24 months, the average return differential reaches as much as 14.5 percent.

This raises the question of whether the starting points at which cost averaging performed better are randomly distributed or concentrated in particular market phases. Figure 3 shows that, for a ten-month investment period, the starting points at which cost averaging outperformed occur in clusters. The prolonged bear market following the bursting of the dot-com bubble and the financial crisis stand out. Investors who committed the entire amount in August 2008 performed around 20 percent worse than those who opted for cost averaging over ten months. However, cost averaging did not prevent an overall loss in that case either. Otherwise, periods in which the lump sum investment was superior predominate, often with a return advantage exceeding 20 percent.

While analysing past performance is straightforward, investors face the difficulty of not knowing whether a prolonged bear market might lie ahead. They can try to predict a bear market using risk indicators. High equity market volatility, measured by 200-day volatility, signals substantial price fluctuations and therefore risk. An inverted yield curve, in which a ten-year US Treasury bond yields less than a three-month Treasury security, is regarded as a potential warning sign of recession.

Figure 4 shows that the lump sum investment was statistically superior both during periods of high equity market volatility and when the US bond market yield curve was inverted. Even with more sophisticated measures, predicting bear markets generally remains unsuccessful.

So far, this analysis has excluded transaction costs and potential interest income on uninvested portions of the capital. Transaction costs reduce the return from cost averaging because each instalment incurs additional costs. Conversely, interest earned on capital that has not yet been invested can increase the return from cost averaging. Historically, transaction costs in Germany averaged 1 percent per transaction. Capital held in current accounts or fixed-term deposits earned an average of 0.3 percent per month. Over a ten-month investment period, the two effects roughly offset each other and therefore barely change the outcome of the return comparison.

Why Investors Still Prefer Cost Averaging

Given these results, the popularity of cost averaging among investors may initially seem surprising. Constantinides (1979) established this statistical finding almost 50 years ago. Why, then, do investors still frequently choose cost averaging?

Behavioural economics offers an answer through the concept of regret aversion (Loomes und Sugden, 1982). Investors want to avoid regretting a decision in hindsight. Anyone who invests their entire capital immediately before even a small market decline may be frustrated by the unfortunate entry point. With cost averaging, this regret is less pronounced because part of the capital can still be invested at lower prices. Falling prices may thus even be perceived as an opportunity. Statman (1995) describes how such psychological considerations can explain the popularity of phased investing. Cost averaging gives investors the feeling that they can better control the risk of entering the market at an unfavourable time.

Another explanation is loss aversion. Investors generally feel losses more acutely than gains of the same magnitude (Kahneman and Tversky, 1979). Those who invest their entire capital at once are immediately exposed to the risk of a loss. With cost averaging, by contrast, only part of the capital is initially exposed to market fluctuations. If prices fall, the absolute loss is smaller.

From the perspective of an individual investor, however, choosing cost averaging is not necessarily irrational. Ludwig von Mises (1949) emphasised that economic action is based on individual values. What is the best decision for one investor needs not be the best for another. Although the lump sum investment is superior in terms of expected returns, cost averaging can reduce perceived risk and concern about entering the market at an unfavourable time.

Returns Are Not Everything: Peace of Mind Comes at a Price

The historical analysis clearly favours the lump sum investment. Investors who commit available capital immediately achieve higher returns on average than those who spread their entry over several months.

Nevertheless, the statistical superiority of the lump sum investment does not translate into a universally applicable recommendation. Investors differ in their perception of risk and their willingness to tolerate temporary losses. Those who reduce their concern about an immediate market decline by investing gradually may derive subjective benefits from doing so and probably sleep better at night as a result. And investors should be allowed a good night’s sleep.

References

Constantinides, George M. (1979): A Note on the Suboptimality of Dollar-Cost Averaging as an Investment Policy. Journal of Financial and Quantitative Analysis 14 (2), 443–450. Available at: doi.org/10.2307/2330513 (accessed 15 October 2026).https://doi.org/10.2307/2330513

Kahneman, Daniel / Tversky, Amos (1979): Prospect Theory: An Analysis of Decision under Risk. Econometrica 47 (2), 263–292. Available at: www.jstor.org/stable/1914185 (accessed 15 October 2026).https://www.jstor.org/stable/1914185

Loomes, Graham / Sugden, Robert (1982): Regret Theory: An Alternative Theory of Rational Choice Under Uncertainty. The Economic Journal 92 (368), 805–824. Available at: doi.org/10.2307/2232669 (accessed 15 October 2026).https://doi.org/10.2307/2232669

Mises, Ludwig von (1949): Human Action: A Treatise on Economics. New Haven: Yale University Press. Available at: mises.org/library/book/human-action (accessed 15 October 2026).https://mises.org/library/book/human-action

Statman, Meir (1995): A Behavioral Framework for Dollar-Cost Averaging. The Journal of Portfolio Management 22 (1), 70–78. Available at: doi.org/10.3905/jpm.1995.409537 (accessed 15 October 2026).https://doi.org/10.3905/jpm.1995.409537