Jan Müller-Dethard, Niklas Reinhardt, and Martin Weber
Review of Finance, Volume 29, Issue 5, September 2025, Pages 1467–1495, https://doi.org/10.1093/rof/rfaf031
It is a long-standing fact that households mostly consume and rarely reinvest dividends. Among representative brokerage clients of one of Germany’s largest banks, we find the opposite: 80% reinvestments, 12% consumption. The figure shows how these results come about. In the month of the dividend payment, only 15% of dividends are reinvested. The remainder is left as cash in the brokerage account and then gradually reinvested into investors’ securities portfolios over the course of the following year. Only few dividends are withdrawn from the brokerage account for consumption.
In further analyses, we find that reinvestment generally occurs on the path of least resistance:
- in months with higher turnover (i.e., when investors trade anyway),
- often through substitutions of securities sales (i.e., investors sell less if they receive more dividends), and
- without additional trades (i.e., investors save themselves a sell order or investors make the same number of buy orders but increase the size of the orders).
Why are our results so vastly different from previous studies that document reinvestment rates between 20% and 30% and consumption rates between 50% and 75% (Baker, Nagel, and Wurgler 2007; Kaustia and Rantapuska 2012; Di Maggio, Kermani, and Majlesi 2020; Bräuer, Hackethal, and Hanspal 2022)? We provide evidence that account structure matters greatly for the question of what investors do with their dividends. Imagine a USD 500 dividend was sent to you as a check in the mail. What would you do with the check? Now, imagine the dividend was directly deposited into your brokerage account. What would you do in this case? Based on mental accounting and default effects, we argue that you would likely treat the dividends as “money for consumption” in the case of the check and as “money for investments” in the case of the brokerage account deposit.
Analyzing the Survey of Consumer Finances, we find that direct deposits have replaced checks as the standard payout method over time. Analyzing the Consumer Expenditure Survey, we find that this matches a strong decline in the dividend consumption rate over time (from 86% in 1988 down to 26% in 2012).
Using evidence from a market overview and a survey that we conduct with retail investors from the US and Germany, we find that our results still leave room for heterogeneity in dividend uses as dividends can be deposited into different types of accounts. While most investors have their dividends deposited into brokerage cash positions (favoring reinvestment, as in our brokerage sample), investors can also have their dividends directly deposited into checking accounts (favoring consumption).
To our knowledge, we are the first to study the account to which investment proceeds are paid as a central element of the choice architecture of brokerage accounts. Our analysis provides important insights to retail investors, financial institutions, and policymakers. At first glance, it might seem convenient for investors to fully integrate all their bank accounts. Yet, our results show that keeping accounts separated could actually help investors achieve higher long-term savings rates and higher stock market participation.